S R
Official journal of the Society for Radiological Protection
P
ACCEPTED MANUSCRIPT
Second follow-up of a German cohort on childhood cancer incidence after exposure to postnatal diagnostic X-ray
To cite this article before publication: Dan Philipp Baaken et al 2019 J. Radiol. Prot. in press https://doi.org/10.1088/1361-6498/ab3506
Manuscript version: Accepted Manuscript
Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process, and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors”
This Accepted Manuscript is @ 2019 Society for Radiological Protection. Published on behalf of SRP by IOP Publishing Limited. All rights reserved ..
During the embargo period (the 12 month period from the publication of the Version of Record of this article), the Accepted Manuscript is fully protected by copyright and cannot be reused or reposted elsewhere.
As the Version of Record of this article is going to be / has been published on a subscription basis, this Accepted Manuscript is available for reuse under a CC BY-NC-ND 3.0 licence after the 12 month embargo period.
After the embargo period, everyone is permitted to use copy and redistribute this article for non-commercial purposes only, provided that they adhere to all the terms of the licence https://creativecommons.org/licences/by-nc-nd/3.0
Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions will likely be required. All third party content is fully copyright protected, unless specifically stated otherwise in the figure caption in the Version of Record.
View the article online for updates and enhancements.
1 2 3 4
Title: Second follow-up of a German cohort on childhood cancer incidence after exposure to postnatal diagnostic X-ray
Authors: Dan Baakenª,1, Gaël P Hammerb, Michael C Seidenbusch“, Karl Schneiderd, Claudia Spixe, Maria Blettnerª, Roman Pokoraª*, Eva Lorenza,f*
a Institute for Medical Biostatistics, Epidemiology and Informatics (IMBEI), Johannes Gutenberg-University, Mainz, Germany
b Registre Morphologique des Tumeurs, Laboratoire National de Santé E.P., Dudelange, Luxembourg
c Institute of Radiation Protection (ISS), Helmholtz Zentrum München - German Research Center for Environmental Health, Neuherberg, Germany
d Department of Radiology, Dr. von Hauner Children’s Medical Hospital, Ludwig- Maximilians-University of Munich, Germany
e German Childhood Cancer Registry, Institute for Medical Biostatistics, Epidemiology and Informatics (IMBEI), Johannes Gutenberg-University, Mainz, Germany
f Bernhard Nocht Institute for Tropical Medicine (BNITM), Hamburg, Germany
* Eva Lorenz and Roman Pokora are joint last authors
1 Address for correspondence
Institute for Medical Biostatistics, Epidemiology and Informatics (IMBEI)
Langenbeckstraße 1
D-55101 Mainz
Germany
Email: dabaaken@uni-mainz.de
Running title: Childhood cancer after diagnostic X-ray: 2nd follow-up
Accepted Ma
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Abstract
Studies on children exposed to ionizing radiation by computed tomography (CT) indicate an increased risk of leukemia and central nervous system (CNS) tumors. Evidence of the risks associated with diagnostic X-ray examinations, the most frequent examination in pediatric radiology, in which the radiation dose is up to 750 times lower compared to CT examinations, is less clear.
This study presents results of the second follow-up for the risk of childhood cancer in a cohort of children (<15 years) with diagnostic X-ray exposure at a large German hospital during 1976-2003 followed for additional 10 years until 2016.
With a latency period of 6 months, 92,998 children contributed 794,549 person-years. The median effective dose was 7 uSv. 100 incident cancer cases were identified: 35 leukemia, 13 lymphomas, 12 CNS tumors, 15 blastomas, 15 sarcomas and 10 other solid tumors, consisting of six germ cells tumors, three thyroid cancers and one adrenocortical carcinoma. For all cancer cases combined the standardized incidence ratio (SIR) was 1.14 (95% confidence interval (CI) 0.93-1.39), for leukemia 1.15 (95% CI 0.63-1.61), for lymphomas 1.03 (95% CI 0.55-1.76), for CNS tumors 0.65 (95% CI 0.34-1.14), for blastomas 1.77 (95% CI 0.91-2.91), for sarcomas 1.28 (95% CI 0.71-2.11) and for other solid tumors 2.38 (95% CI 1.14-4.38). Dose-response analysis using Poisson regression revealed no significant trend for dose groups. Results did not differ substantially with a latency period of 2 years for all cancer entities and 5 years for solid tumors in sensitivity analyses.
Overall, the null results of the first follow-up were confirmed. Although an association between radiation exposure and a risk for certain solid tumors like thyroid cancer is known, the significantly increased SIR in the group of other solid tumors must be critically interpreted in the context of the small number of cases and the very low doses of radiation exposure in this group.
Keywords: Diagnostic X-ray, Childhood cancer, Cohort study, Low dose radiation
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Article Highlights
. This is the second follow-up with 10 more years of follow-up of a cohort of about 100,000 children on the risk of childhood cancer after postnatal diagnostic X-ray.
· In the cohort with very low doses of ionizing radiation (median effective dose 7 uSv) no significantly increased risk for leukemia, lymphomas, CNS tumors, blastomas and sarcomas compared to the general population and no dose-response relationship was found.
· An increased risk for other solid tumors was found, which must be critically interpreted in the light of the small number of cases (n=10) and the very low doses of radiation exposure.
Accepted Manustop
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1. Introduction
Cancer in children is a rare disease, accounting for approximately 1% of all incident malignancies, with an annual incidence of 1.4 per 10,000 live birth in Europe [1]. Nonetheless, cancer is the most common cause of disease-related death in children in industrialized countries [2]. Little is known about the etiology of childhood cancer and the rarity complicates the generation of reliable data on the causes of childhood cancer [3]. One established environmental risk factor is ionizing radiation [3, 4]. Evidence for this mainly comes from studies of atomic bomb survivors in Hiroshima and Nagasaki [5, 6] and a variety of studies on medical exposure to ionizing radiation [7].
In 2009 Hammer and colleagues published results from a large cohort study investigating very low exposure to radiation from diagnostic X-ray examinations, in which they did not observe any increased risk for childhood cancer [8]. Since then several cohort studies on children exposed to computed tomography (CT) with radiation doses up to 750 times higher when comparing, for example, an X-ray examination of the chest to a CT examination of the same region [9], have been published. These recent studies showed an increased risk for leukemia ranging from a relative risk (RR) of 1.19 (95% Confidence Interval (CI) 1.03-1.37) comparing CT versus no CT [10] to a RR of 3.18 (95% CI 1.46-6.94) comparing a cumulative radiation dose to the red bone marrow of <5 mGy versus ≥30 mGy [11]. For brain or CNS tumors a RR of 2.13 (95% CI 1.88-2.41) comparing CT versus no CT [10] to an RR of 3.32 (95% CI 1.84-6.42) comparing <5 mGy versus ≥50 mGy of cumulative radiation dose to the brain [11] was reported.
In terms of radiation protection children are a particularly vulnerable group for mainly two reasons. First, because of a tissue radiosensitivity to radiation-induced cancer compared to adults [12] and second, because of a longer life expectancy in which a cancer can develop [13]. Although doses of ionizing radiation by diagnostic X-ray are far lower compared to newer procedures like CT, they still contribute to the collective medical radiation exposure and thus potentially to the stochastic radiation risk of the population [14]. In addition, diagnostic X-ray examinations are still the most frequent examinations in pediatric radiology involving ionizing radiation [15].
This study represents the second follow-up of a retrospective cohort study published in 2009 by Hammer and colleagues [8] with 10 more years of follow-up of a so far unique cohort and provides an update about the risk of childhood cancer after exposure to very low doses of diagnostic X-ray.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
2. Materials and Methods
2.1 Study Population
This is the second follow-up of a retrospective cohort study published in 2009 by Hammer and colleagues with data from about 100,000 children exposed to postnatal diagnostic X-ray in the years 1976 to 2003 [8]. The study material and methods have been described previously [8]. Briefly, this study was performed by linking pseudonymized cohort data of children exposed to postnatal diagnostic X-ray at the Dr. von Hauner Children’s Hospital, University of Munich (DvHCH) between 1976 and 2003 to the nationwide data of the German Childhood Cancer Registry (GCCR), which started registration in 1980, by stochastic record linkage with the software Merge Toolbox [16]. For this study the stochastic record linkage was done as a complete rerun of the initial record linkage for the first follow-up. Together with an experienced medical documentalist, all potential matches were reviewed in the GCCR.
Besides individual information on basic patient characteristics like sex and age, information on the examined body part, number of X-ray examinations and exposure parameters were available through the electronic documentation system of the DvHCH. Until the introduction of a new Radiological Information System (RIS) in 1998 the referral criteria and diagnosis of examinations, coded according to the International Classification of Disease, 10th revision (ICD-10), were also recorded. Since 1998, this information is not readily available as database files through RIS and therefore not available for this study.
Included patients had to be younger than 14.5 years of age at time of first examination, German residents and without any previous diagnosis of cancer. Patients diagnosed with cancer at the time of their first examination or up to 6 months after diagnosis were excluded. The GCCR has been registering all incident cancer cases in Germany since 1980 for children aged 0 to under 15 years, with a completeness of more than 95% since 1986. Therefore, we had to restrict the cohort to children that were younger than 14.5 years at their first examination, to ensure a minimum latency period of 6 months and additionally had to exclude children older than 15 years at their first examination, because we could not follow them up regarding the occurrence of cancers due to the registration at the GCCR.
The cohort was first followed-up from 1980 to 2006 and now for 10 more years until the end of 2016 while data on exposure from the DvHCH remained the same [8]. Time under risk started 6 months after the initial X-ray examination and ended at date of cancer diagnosis, their 15th birthday, or December 31, 2016, whichever occurred first. We chose a 6-month
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
minimum latency period for the main analysis in accordance with the first follow-up of Hammer and colleagues for a better comparability of the results. In addition, we performed analyses with 2 years latency period for all cancer cases and sensitivity analyses with 5 years latency period for solid cancers. No mortality follow-up was done because of financial reasons. In Germany follow-up for vital status is expensive and mortality rates in children are generally low. The absence of a mortality follow-up could lead to a slight overestimation of the true person-years. Therefore, we reduced the observed person-years by the expected loss due to non-cancer mortality using rates of the general population of Germany in this cohort [17]. In addition, we did sensitivity analysis for children with a potentially high mortality risk, tagged as “high mortality risk” (see Appendix 1).
2.2 Exposure Assessment
Individual radiation doses were previously estimated by Seidenbusch and colleagues using the PAEDOS algorithm developed in the pediatric radiology department at DvHCH. Details on PAEDOS and general exposure assessment are described elsewhere [8, 14]. Briefly, organ doses and whole-body doses were derived from the entrance dose or the dose area product by using conversion coefficients. Conversion coefficients were determined on the basis of the Monte Carlo software PCXMC which was developed by the Finnish Radiation and Nuclear Safety Authority [18]. PCXMC simulates an irradiation to a hermaphrodite mathematical phantom. Simulations were performed for phantoms of different size. The sizes were based on individual data on age at exposure, type of examination, target organ, individual dose area product and other known exposure parameters, e.g. the direction of the beam projection, the patient’s position or the distance from the source to detector [8].
For risk estimation, the individual cumulative effective dose in uSv was used for analyses on all cancer entities combined, lymphomas, blastomas, sarcomas and a combined group of other solid tumors as in the first follow-up [8]. For leukemia and CNS tumors we used individual cumulative tissue doses and organ doses. Red bone marrow dose was used for leukemia and brain dose for CNS tumors. In case of missing doses, we used the median dose of subjects with the same age, sex and year of examination.
2.3 Statistical Analyses
For external analysis, standardized incidence ratio (SIR) with 95% CI were calculated for groups of all cancers (International Classification of Childhood Cancers, 3rd revision (ICCC- 3) I-XII), leukemia (ICCC-3 I), lymphomas (ICCC-3 II), CNS tumors (ICCC-3 III), blastomas (ICCC-3 IV-VII), sarcomas (ICCC-3 VIII-IX) and a combined group of the remaining other
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2 3 4
solid tumors (ICCC-3 X-XII), which are composed of germ cell tumors, trophoblastic tumors, and neoplasms of gonads (ICCC-3 X), other malignant epithelial neoplasms and malignant melanomas (ICCC-3 XI) and other unspecified malignant neoplasms (ICCC-3 XII). In addition, we performed subgroup analyses for Hodgkin lymphomas (ICCC-3 IIa) and Non- Hodgkin lymphomas (except Burkitt lymphoma) (ICCC-3 IIb). We used sex-, age- and calendar year-specific cancer incidence rates of West Germany provided by the GCCR as the reference. Radiation doses were categorized as <1, 1 -< 5, 5 -< 10, 10 -< 25, 25 -< 50, 50 -< 100, 100 -< 250, 250 -< 500 and ≥500 uSv to ensure comparability with the first follow-up. We performed the Cochran-Armitage test for trend to investigate a possible dose-response relationship for these radiation dose categories.
We used Poisson regression modeling to investigate a possible dose-response relationship between doses of X-rays and the risk of childhood cancer. Incidence rate ratios (IRR) were adjusted for sex and age at diagnosis and 95% CI were calculated. IRR were calculated for all cancers, leukemia, lymphomas, CNS tumors, blastomas, sarcomas and other solid tumors by dose categorized as 0 -< 10 uSv as reference category, 10 -< 50 uSv and ≥50 uSv. Again, these dose categories were chosen to make the results comparable to the first follow-up.
To ensure comparability with previously published case-control studies additionally the number of examinations categorized as 1, 2 and 23 X-ray procedures was used as a dose proxy. The Cochrane-Armitage test for trend was also performed for the number of examinations. For all analyses, a latency period of 6 months was used. In addition, we used a latency period of 2 years for all cancer entities and of 5 years for solid tumors for sensitivity analyses and we performed subgroup analyses on patients who were exposed within the first 100 days after birth, to investigate the influence of X-ray examinations very early in life on cancer incidence in childhood. Because the cohort included some heterogeneous subgroups, the following groups of patients based on available diagnostic or exposure information were defined to be able to differentiate them in analyses: “highly exposed”, “high mortality risk”, “elevated incidence risk”, “premature”. Children were labeled as “highly exposed” if they had a recorded CT or contrast media examination. Exposures from these examinations have not been reconstructed. These examinations, which were performed on 3.7% of the cohort, were documented. Reconstruction of doses were however not possible within this project due to the application of more complex and time-consuming methods. Children were assigned to the group of “high mortality risk” if they had a serious disease such as AIDS or hydrocephalus.
Children were marked as “elevated cancer risk” if they were diagnosed with a syndrome that
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2 3 4
is known to be associated with an increased cancer incidence like Down’s syndrome. Premature babies have an increased mortality risk and are repeatedly exposed to X-ray examinations to check the development of the lungs. Children were tagged as “premature” when this was as an indication or diagnosis for an X-ray examination. A complete list of labels of ICD-10 blocks and disease for the tags “high mortality risk” and “elevated cancer risk” and the corresponding ICD-10 codes can be found in the Supplementary material (Appendix 1) and in the publication of Hammer and colleagues [8].
3. Results
3.1 Study Population
In total, the cohort included information on 105,847 children, who had at least one postnatal diagnostic X-ray examination at the DvHCH. Of these, 12,849 children had to be excluded from analyses: 9,757 children were 14.5 or older at their first X-ray examination, and 1,547 were 15 years or older at the start of follow-up. 1,363 children had a prevalent cancer at time of the first X-ray examination or within 6 months after the first examination or before beginning of the follow-up, 176 children had inconsistent data on date of birth and examination and six children had an examination after 2003. 41 fewer children were excluded from the study compared to the first follow-up because of a rerun of the probabilistic record linkage and possibly changes in the dynamic registry of the GCCR in terms of name spelling or date of birth.
The cohort consists of 92,998 children: 50,010 boys, 41,467 girls and 1,521 children of unknown sex. During the analysis, we discovered a small overestimation of the total person- years in the previous analysis (4.19% of the true person-years) which led to an overestimation of the mean follow-up time of 0.3 years. Compared to the first follow-up until 2006 with corrected person-years of 696,970 and corrected mean follow-up of 7.5 years, the extended follow-up until 2016 increased the total number of person-years by 97,624 to 794,594 person- years and the mean follow-up time by 1.0 year to a mean follow-up time of 8.5 years in this study [8].
Overall 100 incident cancer cases (60 boys and 40 girls) were found by record linkage to the GCCR between 1980 and 2016 using a latency period of 6 months (Table 1). The group of other solid tumors (ICCC-3 X-XII) includes 4 cases of intracranial and intraspinal germ cell tumors (ICCC-3 Xa), 3 cases of thyroid carcinomas (ICCC-3 XIb) and one case each of malignant extracranial and extragonadal germ cell tumors (ICCC-3 Xb), malignant gonodal germ cell tumors (ICCC-3 Xc) and adrenocortical carcinomas (ICCC-3 XIa). When using a
8
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
latency period of 2 years 71 cancer cases were observed (Table 1) and 39 cases when applying a latency period of 5 years (Table 1). 20 of the incident cancer cases were tagged as “patients with high mortality risk”, 5 were tagged as “highly exposed” and one case was tagged as “elevated incidence risk”. There was no case that was tagged as “premature”. The most frequent cancer was acute lymphoblastic leukemia with 24 cases (Table 1). The mean age at diagnosis was 7.4 years (range 1-14, standard deviation 4.2, median 7,3). Infant cancer cases were not observed.
When restricting our analyses to children exposed to diagnostic radiation in the first 100 days after birth, we identified 11,403 children of whom 21 were diagnosed with cancer during childhood (8 leukemia, 2 lymphomas, 4 CNS tumors, 4 blastomas, 3 sarcomas).
3.2 Exposure
Most children (59%) had only one radiographic examination at the DvHCH. About 19% of the cohort had two examinations and 22% had three or more examinations. The mean age at first examination at the DvHCH was 5.6 years. About 22% of the cohort had their first examination at 0 years of age, 10% at one year of age, 21% from 2 to 4 years, about 27% at 5 to 9 years and 20% in the age group of 10-14 years. Of the 100 incident cancer cases, 59 children had only one radiographic examination. A further 16 patients had two, and 25 children had three or more examinations. 42 of all incident cancer cases had their first examination at 0 years of age, while 15 cases had their first examination at one year of age, 24 between 2 and 4 years, 14 between 5 and 9 years and 5 cases in the age group of 10-14 years (Table 1). Individual cumulative effective doses ranged from 0 to 343,400 µSv. The median cumulative effective dose in the cohort was 7 uSv and the mean cumulative effective dose was 134.7 uSv. For the cancer cases, the median cumulative effective dose was 20 uSv and the mean cumulative effective dose was 114.5 uSv.
3.3 Statistical Analysis
The SIR for all incident cancers was 1.14 (95% CI 0.93-1.39). The SIR for leukemia, lymphomas (in general and for Hodgkin lymphomas and Non-Hodgkin lymphomas), CNS tumors, blastomas and sarcomas did not differ significantly from 1.0 (Table 2). In the group of other solid tumors, the SIR was 2.38 (95% CI 1.14-4.38) (Table 2). For all cancers combined, the SIR generally did not differ between sexes (Table 2), and in analyses for dose- categories, no single SIR was significant and no trend with increasing dose was observed (Table 2). The same applies to the number of examinations as a dose proxy (Table 2).
Subgroup analyses were carried out for patients tagged as “high mortality risk” with 120,552
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
55 56 57 58 59 60
person-years and patients tagged as “highly exposed” with 28,831 person-years (Table 2). SIR were not significantly raised for these subgroups. Other subgroups had too few cases for analyses.
No overall trend of increasing cancer risk with increasing dose was observed in the IRR analyses for all cancers combined, leukemia, lymphomas, CNS tumors, blastomas, sarcomas and other solid tumors (Table 3 and Table 4). In IRR analyses for children excluding patients tagged as “highly exposed”, no association was observed for dose categories of 10- < 50 uSv and ≥50 uSv compared to the reference category (Table 3 and Table 4). There were too few cases in the groups labeled as “high mortality risk” “elevated cancer risk” and “premature”, so no regression model could be fit. In sensitivity analyses with the number of examinations as a dose proxy, a dose-response relationship was not observed either.
No increased risk could be observed in an additional IRR analysis based on the 21 cancer cases with exposure to diagnostic radiation within the first 100 days after birth adjusted for sex and age at diagnosis in exposure category of 10-50 uSv (IRR 1.04 CI 0.32-3.36) and of ≥50 µSv (IRR 0.79 CI 0.24-2.61) compared to the reference group of <10 uSv.
A latency period of 2 years for all cancer entities observed yielded 71 incident cancer cases and 680,825 person-years in the cohort. For all cancers combined, the SIR was lower compared to a latency period of 6 months. The same applies to leukemia, lymphomas (in general, and Hodgkin lymphomas and Non-Hodgkin lymphomas), and CNS tumors. In the groups of blastomas, sarcomas and other solid tumors, the SIR increased, with other solid tumors remaining significant (Table 2). In sensitivity analyses with a latency period of 5 years for CNS tumors, sarcomas, and other solid tumors the SIR for other solid tumors remained significantly raised and the SIR for blastomas became significantly increased (SIR 3.57 95% CI 1.44-7.36) (Appendix 2). Results for IRR analyses with 2-year latency period again showed no dose-response relationship (Table 3), and the same applies for analyses using a 5- year latency period on blastomas, sarcomas and other solid tumors (Appendix 3). However, it was not possible to perform a dose-response analyses on CNS tumors due to the low number of cases (n=7).
4. Discussion
Our study is the second follow-up, adding further 10 years of follow-up and additional analyses with organ doses, extended latency periods, and subgroups of cancer entities, of a cohort of children with postnatal diagnostic X-ray and the risk of childhood cancer. Overall,
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
794,549 person-years with a mean of 8.5 years follow-up were observed with 100 incident childhood cancer cases under the age of 15 compared to 87 cases after the first follow-up in a cohort with an overall very low median cumulative effective dose of 7 uSv. The relatively small increase in the mean follow-up time of 14% is due to the fact that many children from the cohort already reached the maximum included age of 14.5 years in the first follow-up until 2006.
Since 2009, three case-control studies [19-21] on postnatal diagnostic X-rays cancer risk have been published. The results of our study are in line with the results of a case-control study of Rajaraman and colleagues, who observed an odds ratio (OR) of 1.16 (95% CI 0.83-1.16) for all cancers in children exposed to diagnostic X-rays in the first 100 days after birth [20]. They also found no statistically significantly increased risks for leukemia or lymphomas and a reduced risk for CNS tumors. In our additional analyses, which were limited to children with diagnostic radiation in the first 100 days after birth we also did not observe an increased risk for all cancers in children. The distribution of the occurring 21 cancer cases of children exposed in the first 100 days after birth was comparable to the distribution of all cancer cases in the cohort. The number of cases in our study was too small for further stratified analyses for cancer entities of this group. In a study of Milne and colleagues focusing on childhood brain tumors, no evidence of a positive association with diagnostic X-ray in childhood had been observed [21]. Bartley and colleagues suggest that postnatal diagnostic X-rays are associated with an increased risk for acute lymphoblastic leukemia (ALL) (OR 1.85 CI 1.12- 2.79) when exposed to three or more examinations, but no association for acute myeloid leukemia (AML) was observed [19]. When restricting our analysis on those two subgroups of leukemia patients and comparing children with one postnatal diagnostic X-ray with children with three or more diagnostic X-rays we could not observe an increased risk for ALL (IRR 0.31 95% CI 0.07-1.35) with very low numbers of cases (10 cases with one diagnostic X-ray, 2 cases with three or more diagnostic X-rays). For AML we found an IRR of 8.50 (95% CI 0.88-81.77), which should be interpreted with caution because of the wide CI resulting from a very small number of cases (1 case with one diagnostic X-ray, 3 cases with three or more diagnostic X-rays).
Overall the risk estimates of this extended follow-up are in line with the null results of the first follow-up, which one would expect with the very low effective doses in this cohort. However, the observed increased risk for other solid tumors for all latency periods (6 months, 2 years, 5 years) must be interpreted with caution since the number of cases for this group
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(n=10), mainly consisting of germ cell tumors (n=6) and thyroid carcinomas (n=3) is very small. Although an association between radiation exposure and the risk for certain solid tumors like thyroid cancer is well documented [22, 23] our findings could be due to chance because of the small number of cases. For this very reason no more stratified analyses with even smaller subgroups were performed. The individual cumulative effective doses in this cohort, in general, are very low. Of the 10 cases of other solid tumors 8 had a cumulative exposure of <10 uSv effective dose and two had between 10 -< 50 uSv. At these low doses a radiation induced significant increased effect would not be expected. The same limitation in terms of the interpretation applies to the increased SIR for blastomas in sensitivity analyses with a 5-year latency period (n=7). Although the cumulative exposure in this group was higher. Four children had a cumulative exposure of ≥50 uSv effective dose, two children had between 10 -< 50 uSv and one children had <10 uSv.
In the discussion of the recent CT-studies [10, 11, 24-26] and the observed relationship between CT and increased risks for leukemia and brain tumors, the issue of confounding by indication and reverse causation came up since CT is a commonly used technique for diagnosing cancer, particularly solid cancers [27-29]. In our study information on the indication for the X-ray examination and the resulting diagnosis were available for X-rays until 1998. Any patients with an indication of cancer as the reason for the examination were excluded from the cohort. Nevertheless, we cannot completely rule out that certain cancer susceptibility syndromes like the Li-Fraumeni syndrome for leukemia and brain tumors or the Noonan syndrome for leukemia [30] caused an increased risk for childhood cancer or lead to additional CT scans that would confound our results. However, we did not observe an increased risk for leukemia or CNS tumors in our study.
The main strength of this study is avoiding a possible recall bias compared to those case- control studies on childhood cancer after diagnostic X-ray, which use self-reported history of X-rays from the parents [19, 21]. Further strengths of our study are: Absorbed doses for organs and tissues were reconstructed individually from medical records at the DvHCH [8] and the identification of incident cancer cases was performed via a registry with a completeness of above 95% of incident cancer cases in Germany and a high data quality [31].
Limitations of this study are: Data on exposure was exclusively collected at the radiology department of the DvHCH and patients may have had further X-rays examinations elsewhere. In addition, children could have been exposed to further X-ray examinations before 1976, about which we had no information. Both limitations lead to an underestimation of the
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2 3 4
exposure. While CT or contrast media examinations were recorded and tagged as “highly exposed”, the corresponding doses were not reconstructed, which leads to an additional underestimation of the individual radiation dose [8]. With the introduction of the RIS at the DvHCH in 1998, information on indication and radiological examination were no longer abstracted from (often handwritten) medical files so children, who were examined to clarify possibly cancer-related symptoms, could not be identified and excluded after 1997. This could lead to a possible overestimation of the observed number of incident cancer cases in childhood. Nevertheless, the risk for a possible overestimation should be reduced by the latency period, especially by the 2-year latency period and the 5-year period in sensitivity analyses for solid tumors. In addition, we did not exclude patients born before the beginning of the cancer registration at the GCCR in 1980. Therefore, we cannot completely rule out that children born before 1980 had a first incident cancer before 1980 and had a second neoplasm before the age of 15 after 1980, which was documented as a first incident cancer case. Although the number of these cases should be small this could lead to an overestimation of the observed number of incident cancer cases. Results of additional analyses with children born before 1980 excluded were generally in line with the results of the complete cohort and did not change the conclusion (data not shown).
Residual confounding is a further limitation of this study. Possible confounding factors could be in utero exposure to ionizing radiation [32], maternal alcohol consumption [33] and smoking [34]. In addition, when calculating SIR, the observed cases in the cohort were compared to the general population that can be exposed to diagnostic X-ray examinations too. An exposure free comparison group for the external comparison would have been more informative, but was unavailable.
5. Conclusion
The analysis of Hammer and colleagues and this extended follow-up analysis suggest that exposure to very low doses of postnatal conventional diagnostic X-rays (median effective dose of 7 uSv) in childhood is not associated with an increased risk for childhood cancer under 15 years. There was no observed increased risk for leukemia, lymphomas, CNS tumors, blastomas and sarcomas compared to the general population and no dose-response relationship. The increased risk for other solid tumors must be critically interpreted in the light of the small number of cases and the very low doses of radiation exposure.
The latest published Commentary No. 27 of the National Council on Radiation Protection and Measurements (NCRP) concludes that recent epidemiologic studies support the use of the
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
linear no-threshold (LNT) model for the purposes of radiological protection, which is in accordance with the older judgments by other national and international scientific committees [35]. In the light of this commentary of the NCRP on LNT and for reasons of radiation protection [36] an extension of this cohort with further follow-up of cancer incidence into adult age and additional dosimetry including CT could be a future goal, which would provide more insight into the risk of cancer in a cohort with very low doses of ionizing radiation due to postnatal diagnostic X-ray examination during childhood.
Acknowledgments
The authors thank the staff of the German Childhood Cancer Registry especially C. Bremensdorfer, who helped with the clerical review and T. Ziegler, who helped with the matching.
Accepted Manuscript
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
References
[1] Kaatsch P 2010 Epidemiology of childhood cancer Cancer treatment reviews 36 277- 85
[2] Buka I, Koranteng S and Osornio Vargas A R 2007 Trends in childhood cancer incidence: review of environmental linkages Pediatr Clin North Am 54 177-203
[3] Spector L G, Pankratz N and Marcotte E L 2015 Genetic and nongenetic risk factors for childhood cancer Pediatr Clin North Am 62 11-25
[4] Belson M, Kingsley B and Holmes A 2007 Risk Factors for Acute Leukemia in Children: A Review Environ Health Perspect 115 138-45
[5] Walsh L and Kaiser J C 2011 Multi-model inference of adult and childhood leukaemia excess relative risks based on the Japanese A-bomb survivors mortality data (1950- 2000) Radiat Environ Biophys 50 21-35
[6] Preston D L, Cullings H, Suyama A, Funamoto S, Nishi N, Soda M, Mabuchi K, Kodama K, Kasagi F and Shore R E 2008 Solid cancer incidence in atomic bomb survivors exposed in utero or as young children J Natl Cancer Inst 100 428-36
[7] Wakeford R 2013 The risk of childhood leukaemia following exposure to ionising radiation — a review Journal of radiological protection : official journal of the Society for Radiological Protection 33 1-25
[8] Hammer G P, Seidenbusch M C, Schneider K, Regulla D F, Zeeb H, Spix C and Blettner M 2009 A cohort study of childhood cancer incidence after postnatal diagnostic X-ray exposure Radiat Res 171 504-12
[9] Linet M S, Kim K P and Rajaraman P 2009 Children’s exposure to diagnostic medical radiation and cancer risk: epidemiologic and dosimetric considerations Pediatr Radiol 39 Suppl 1 S4-26
[10] Mathews J D, Forsythe A V, Brady Z, Butler M W, Goergen S K, Byrnes G B, Giles G G, Wallace A B, Anderson P R, Guiver TA, McGale P, Cain T M, Dowty J G, Bickerstaffe A C and Darby S C 2013 Cancer risk in 680,000 people exposed to computed tomography scans in childhood or adolescence: data linkage study of 11 million Australians BMJ 346 f2360
[11] Pearce M S, Salotti J A, Little M P, McHugh K, Lee C, Kim K P, Howe N L, Ronckers C M, Rajaraman P, Sir Craft A W, Parker L and Berrington de Gonzalez A 2012 Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study Lancet 380 499-505
[12] Alzen G and Benz-Bohm G 2011 Radiation Protection in Pediatric Radiology Deutsches Ärzteblatt International 108 407-14
[13] Shah N B and Platt S L 2008 ALARA: is there a cause for alarm? Reducing radiation risks from computed tomography scanning in children Current opinion in pediatrics 20 243-7
[14] Seidenbusch M C and Schneider K 2008 Zur Strahlenexposition von Kindern in der pädiatrischen Radiologie Fortschr Röntgenstr 180 1082-103
[15] Brix G, Nekolla E and Griebel J 2005 Radiation exposure of patients from diagnostic and interventional X-ray procedures. Facts, assessment and trends Radiologe 45 340-9
[16] Schnell R, Bachteler, T., Reiher, T. 2005 Ein Record-Linkage-Programm fuer die empirische Sozialforschung ZA-Information 56 93-103
[17] Becher H and Winkler V 2017 Estimating the standardized incidence ratio (SIR) with incomplete follow-up data BMC Med Res Methodol 17 55
[18] Tapiovaara M, Lakkisto M and Sercvomaa A 1997 PCXMC: A PC-based Monte Carlo program for calculating patient doses in medical x-ray examinations Report No. STUK- A139, Finnish Centre for Radiation and Nuclear Safety, Helsinki
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
[19] Bartley K, Metayer C, Selvin S, Ducore J and Buffler P 2010 Diagnostic X-rays and risk of childhood leukaemia Int J Epidemiol 39 1628-37
[20] Rajaraman P, Simpson J, Neta G, Berrington de Gonzalez A, Ansell P, Linet M S, Ron E and Roman E 2011 Early life exposure to diagnostic radiation and ultrasound scans and risk of childhood cancer: case-control study BMJ 342 d472
[21] Milne E, Greenop K R, Fritschi L, Attia J, Bailey H D, Scott R J, Ashton L J, Smibert E and Armstrong B K 2014 Childhood and parental diagnostic radiological procedures and risk of childhood brain tumors Cancer Causes Control 25 375-83
[22] Iglesias M L, Schmidt A, Ghuzlan A A, Lacroix L, Vathaire F, Chevillard S and Schlumberger M 2017 Radiation exposure and thyroid cancer: a review Archives of endocrinology and metabolism 61 180-7
[23] Han M A and Kim J H 2018 Diagnostic X-Ray Exposure and Thyroid Cancer Risk: Systematic Review and Meta-Analysis Thyroid : official journal of the American Thyroid Association 28 220-8
[24] Berrington de Gonzalez A, Salotti J A, McHugh K, Little M P, Harbron R W, Lee C, Ntowe E, Braganza M Z, Parker L, Rajaraman P, Stiller C, Stewart D R, Craft A W and Pearce M S 2016 Relationship between paediatric CT scans and subsequent risk of leukaemia and brain tumours: assessment of the impact of underlying conditions Br J Cancer 114 388-94
[25] Huang W Y, Muo C H, Lin C Y, Jen Y M, Yang M H, Lin J C, Sung F C and Kao CH 2014 Paediatric head CT scan and subsequent risk of malignancy and benign brain tumour: a nation-wide population-based cohort study Br J Cancer 110 2354-60
[26] Krille L, Dreger S, Schindel R, Albrecht T, Asmussen M, Barkhausen J, Berthold J D, Chavan A, Claussen C, Forsting M, Gianicolo E A L, Jablonka K, Jahnen A, Langer M, Laniado M, Lotz J, Mentzel H J, Queisser-Wahrendorf A, Rompel O, Schlick I, Schneider K, Schumacher M, Seidenbusch M, Spix C, Spors B, Staatz G, Vogl T, Wagner J, Weisser G, Zeeb H and Blettner M 2015 Risk of cancer incidence before the age of 15 years after exposure to ionising radiation from computed tomography: results from a German cohort study Radiat Environ Biophys 54 1-12
[27] Boice J D, Jr. 2015 Radiation epidemiology and recent paediatric computed tomography studies Ann ICRP 44 236-48
[28] Journy N, Rehel J L, Ducou Le Pointe H, Lee C, Brisse H, Chateil J F, Caer-Lorho S, Laurier D and Bernier M O 2015 Are the studies on cancer risk from CT scans biased by indication? Elements of answer from a large-scale cohort study in France Br J Cancer 112 185-93
[29] Walsh L, Shore R, Auvinen A, Jung T and Wakeford R 2014 Risks from CT scans — what do recent studies tell us? Journal of radiological protection : official journal of the Society for Radiological Protection 34 E1-5
[30] Meulepas J M, Ronckers C M, Merks J, Weijerman M E, Lubin J H and Hauptmann M 2016 Confounding of the Association between Radiation Exposure from CT Scans and Risk of Leukemia and Brain Tumors by Cancer Susceptibility Syndromes Cancer Epidemiol Biomarkers Prev 25 114-26
[31] Kaatsch P 2004 German Childhood Cancer Registry and its favorable setting Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 47 437-43
[32] Linet M S, Kim K P and Rajaraman P 2009 Children’s exposure to diagnostic medical radiation and cancer risk: epidemiologic and dosimetric considerations Pediatric Radiology 39 4-26
[33] van Duijn C M, van Steensel-Moll H A, Coebergh J W and van Zanen G E 1994 Risk factors for childhood acute non-lymphocytic leukemia: an association with maternal alcohol consumption during pregnancy? Cancer Epidemiol Biomarkers Prev 3 457-60
V
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
[34] Rumrich IK, Viluksela M, Vahakangas K, Gissler M, Surcel H M and Hanninen O 2016 Maternal Smoking and the Risk of Cancer in Early Life - A Meta-Analysis PLoS One 11 e0165040
[35] NCRP 2018 Implications of Recent Epidemiologic Studies for the Linear Nonthreshold Model and Radiation Protection Commentary No. 27 (Bethesda, MD: National Council on Radiation Protection and Measurements)
[36] Shore R E, Beck H L, Boice J D, Caffrey E A, Davis S, Grogan H A, Mettler F A, Preston R J, Till J E, Wakeford R, Walsh L and Dauer L T 2018 Implications of recent epidemiologic studies for the linear nonthreshold model and radiation protection Journal of radiological protection : official journal of the Society for Radiological Protection 38 1217-33
[37] Steliarova-Foucher E, Stiller C, Lacour B and Kaatsch P 2005 International Classification of Childhood Cancer, third edition Cancer 103 1457-67
Accepted Manusers
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
S
| Diagnosis | Sex Age at diagnosis of cancer Total ICCC-3ª Boys Girls 0 1-4 5-9 10-14 6-month 2-year 5-year latency latency latency | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| All cancers | I-XII | 60 | 40 | 0 | 34 | 32 | 34 | 100 | 71 | 39 |
| Leukemias, myeloproliferative | I | 19 | 16 | 0 | 15 | 10 | 10 | 35 | 24 | 8 |
| diseases, and myelodysplastic | ||||||||||
| diseases | ||||||||||
| Lymphocytic Leukemia | Ia | 14 | 10 | 0 | 12 | 5 | 7 | 24 | 16 | 4 |
| Acute myeloid Leukemia | Ib | 2 | 2 | 0 | 0 | 2 | 2 | 4 | 4 | 2 |
| other Leukemias | Ic-Ie | 3 | 4 | 0 | 3 | 2 | 2 | 7 | 4 | 2 |
| Lymphoma | II | 12 | 1 | 0 | 5 | 6 | 2 | 13 | 7 | 3 |
| Hodgkin lymphoma | IIa | 4 | 0 | 0 | 1 | 2 | 1 | 4 | 1 | 0 |
| Non-Hodgkin lymphoma | IIb | 7 | 1 | 0 | 4 | 3 | 1 | 8 | 5 | 3 |
| (except Burkitt lymphoma) | ||||||||||
| CNS tumors | III | 8 | 4 | 0 | 3 | 5 | 4 | 12 | 8 | 7 |
| Neuroblastoma and other peripheral | IV | 1 | 3 | 0 | 2 | 2 | 0 | 4 | 3 | 2 |
| nervous cell tumors | ||||||||||
| Retinoblastoma | V | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
| Renal tumors | VI | 5 | 3 | 0 | 3 | 5 | 0 | 8 | 6 | 4 |
| Hepatic tumors | VII | 2 | 0 | 0 | 1 | 0 | 1 | 2 | 1 | 1 |
| Malignant bone tumors | VIII | 3 | 0 | 0 | 0 | 0 | 3 | 3 | 3 | 3 |
| Soft tissue and other extraosseous | IX | 5 | 7 | 0 | 4 | 3 | 5 | 12 | 10 | 4 |
| sarcomas | ||||||||||
| Germ cell tumors, trophoblastic | X | 3 | 3 | 0 | 0 | 0 | 6 | 6 | 5 | 4 |
| tumors, and neoplasms of gonads | ||||||||||
| Other malignant epithelial neoplasms | XI | 1 | 3 | 0 | 0 | 1 | 3 | 4 | 4 | 3 |
| and malignant melanomas | ||||||||||
| Other and unspecified malignant | XII | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| neoplasms | ||||||||||
L
18
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
ª ICCC-3: International Classification of Childhood Cancers, 3rd revision [37].
Accepted Manus®
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
1 2 3 4 5
Table 2: Standardized incidence ratios (SIR) of incident cancer cases in a cohort of children who underwent diagnostic X-ray procedures at Dr. von Hauner Children’s Hospital, Munich, Germany, in the period 1976-2003 by cancer type, sex, assigned tags, cumulative effective dose and number of examinations
X
| 6-month latency period | 2-year latency period | |||||||
|---|---|---|---|---|---|---|---|---|
| 0ª | Eª | SIRª | 95% CIª | 0ª | Eª | SIRª | 95% CIª | |
| All cancers | 10 | 88.0 | 0.93-1.39 | 71 | 72.7 | 0.98 | 0.76-1.23 | |
| 0 | 1.14 | |||||||
| Leukemia | 35 | 30.5 | 1.15 | 0.80-1.61 | 24 | 24.3 | 0.99 | 0.63-1.47 |
| Lymphoblastic | 24 | 24.2 | 0.99 | 0.63-1.47 | 16 | 19.2 | 0.83 | 0.48-1.36 |
| leukemia | ||||||||
| Acute myeloid | 4 | 4.2 | 0.96 | 0.26-2.45 | 4 | 3.5 | 1.15 | 0.31-2.94 |
| leukemia | ||||||||
| Lymphomas | 13 | 12.7 | 1.03 | 0.55-1.76 | 7 | 11.4 | 0.61 | 0.25-1.26 |
| Hodgkin lymphoma | 4 | 6.6 | 0.60 | 0.16-1.54 | 1 | 4.8 | 0.21 | 0.01-1.16 |
| Non-Hodgkin | ||||||||
| lymphoma | 8 | 5.1 | 1.56 | 0.57-3.08 | 5 | 3.7 | 1.35 | 0.44-3.16 |
| CNS tumors | 12 | 18.4 | 0.65 | 0.34-1.14 | 8 | 15.7 | 0.51 | 0.22-1.00 |
| Blastomas | 15 | 8.5 | 1.77 | 0.91-2.91 | 10 | 5.3 | 1.89 | 0.91-3.47 |
| Sarcomas | 15 | 11.8 | 1.28 | 0.71-2.11 | 13 | 8.4 | 1.54 | 0.82-2.64 |
| Other solid tumors | 10 | 4.2 | 2.38 | 1.14-4.38 | 9 | 2.7 | 3.32 | 1.52-6.31 |
| Sex (all cancers) | ||||||||
| Boys | 60 | 52.5 | 1.14 | 0.87-1.47 | 42 | 43.7 | 0.96 | 0.69-1.30 |
| Girls | 40 | 35.1 | 1.00 | 0.81-1.55 | 29 | 29.0 | 1.00 | 0.67-1.43 |
| Patients tagged as having | ||||||||
| "high mortality risk"b,c | ||||||||
| No | 80 | 73.0 | 1.10 | 0.87-1.36 | 55 | 60.1 | 0.91 | 0.69-1.19 |
| Yes | 20 | 14.7 | 1.36 | 0.83-2.10 | 16 | 12.6 | 1.27 | 0.72-2.06 |
| "Highly exposed“ patientsb | ||||||||
| No | 95 | 84.2 | 1.13 | 0.91-1.38 | 69 | 69.7 | 0.99 | 0.77-1.25 |
| Yes | 5 | 3.5 | 1.43 | 0.46-3.34 | 2 | 3.0 | 0.67 | 0.08-2.40 |
| Effective dose category | ||||||||
| (uSv) (all cancers) | ||||||||
| <1 | 16 | 18.9 | 0.85 | 0.49-1.38 | 7 | 14.7 | 0.48 | 0.19-0.98 |
| 1 -< 5 | 13 | 11.4 | 1.14 | 0.61-1.95 | 12 | 9.3 | 1.29 | 0.67-2.25 |
| 5 -< 10 | 17 | 12.0 | 1.42 | 0.83-2.27 | 11 | 10.0 | 1.10 | 0.55-1.97 |
| 10 -< 25 | 15 | 13.7 | 1.09 | 0.61-1.80 | 10 | 11.7 | 0.86 | 0.41-1.58 |
| 25 -< 50 | 15 | 8.4 | 1.79 | 1.00-2.95 | 11 | 7.2 | 1.54 | 0.77-2.75 |
| 50 -< 100 | 6 | 7.3 | 0.82 | 0.30-1.79 | 4 | 6.3 | 0.64 | 0.17-1.63 |
| 100 -< 250 | 7 | 7.7 | 0.91 | 0.37-1.87 | 6 | 6.6 | 0.91 | 0.33-1.99 |
| 250 -< 500 | 5 | 4.0 | 1.26 | 0.41-2.93 | 5 | 3.4 | 1.47 | 0.48-3.44 |
| ccepte ≥500 | 6 | 4.3 | 1.40 | 0.51-3.04 | 5 | 3.7 | 1.36 | 0.44-3.18 |
| Trend test: P value | 0.31 | 0.12 | ||||||
| Number of examinations | ||||||||
| (all cancers) | ||||||||
| 1 | 59 | 53.0 | 1.11 | 0.85-1.44 | 38 | 44.1 | 0.86 | 0.61-1.18 |
| 2 | 16 | 16.1 | 1.00 | 0.57-1.62 | 13 | 13.3 | 0.97 | 0.52-1.67 |
| ≥3 | 25 | 18.6 | 1.35 | 0.87-1.99 | 20 | 15.4 | 1.30 | 0.80-2.01 |
| Trend test: P value | 0.33 | 0.16 | ||||||
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ª O: observed cases, E: expected cases, SIR: standardized incidence ratio, CI: confidence interval.
b See methods section for a definition.
” excluding patients labelled as “highly exposed”.
Accepted Manuscript
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
| Group | Dose catego ryb (uSv) | All cancers (I-XII) | Leukemia (I) | Lymphoma (III-XII) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | ||
| 6-month latency period | ||||||||||
| All patients | 0 -< 10 | 46 | Reference | 21 | Reference | 6 | Reference | |||
| 10 -< 50 | 30 | 1.10 | 0.69-1.75 | 8 | 1.15 | 0.51-2.60 | 3 | 1.03 | 0.24-4.32 | |
| ≥50 | 24 | 0.84 | 0.51-1.37 | 6 | 0.93 | 0.37-2.30 | 4 | 1.38 | 0.37-5.20 | |
| All patients | 0 -< 10 | 46 | Reference | 20 | Reference | 6 | Reference | |||
| without | 10 -< 50 | 27 | 1.03 | 0.64-1.66 | 8 | 1.24 | 0.55-2.82 | 3 | 0.79 | 0.20-3.18 |
| "highly | ≥50 | 22 | 0.80 | 0.48-1.33 | 6 | 1.00 | 0.40-2.48 | 3 | 0.79 | 0.20-3.17 |
| exposed"" | ||||||||||
| 2-year latency period | ||||||||||
| All patients | 0 -< 10 | 30 | Reference | 12 | Reference | 3 | Reference | |||
| 10 -< 50 | 21 | 1.26 | 0.72-2.21 | 6 | 1.45 | 0.54-3.86 | 2 | 1.18 | 0.18-7.53 | |
| ≥50 | 20 | 1.12 | 0.64-1.98 | 6 | 1.51 | 0.57-4.03 | 2 | 1.17 | 0.18-7.55 | |
| All patients | 0 -< 10 | 30 | Reference | 12 | Reference | 3 | Reference | |||
| without | 10 -< 50 | 19 | 1.17 | 0.65-2.08 | 6 | 1.48 | 0.55-3.94 | 2 | 0.99 | 0.16-5.98 |
| "highly exposed"c | ≥50 | 20 | 1.16 | 0.66-2.05 | 6 | 1.55 | 0.58-4.13 | 2 | 0.99 | 0.13-5.95 |
ª IRR: incidence rate ratio, CI: 95% Wald confidence interval.
Effective dose for all cancers and lymphoma, tissue dose for leukemia (red bone marrow dose).
” See methods section for a definition.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
1 2 3 4
| Group | Dose categoryb (uSv) | CNS tumors (III) | Blastomas (IV-VII) | Sarcomas (VIII-IX) | Other solid tumors (X- XII) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Case s | IRRª 95% CIª | Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | ||
| 6-month latency period | ||||||||||||
| All | 0 -< 10 | 11 | Reference | 6 | Reference | 6 | Reference | 8 | Reference | |||
| patients | 10 -< 50 | 0 | 4 | 1.00 | 0.28-3.57 | 7 | 2.17 | 0.72-6.51 | 2 | 0.57 | 0.12-2.71 | |
| ≥50 | 1 | 0.95 0.12-7.40 | 5 | 1.21 | 0.37-3.98 | 2 | 0.57 | 0.12-2.84 | 0 | |||
| All | 0 -< 10 | 11 | Reference | 6 | Reference | 6 | Reference | 8 | Reference | |||
| patients | 10 -< 50 | 0 | 3 | 0.77 | 0.19-3.09 | 6 | 1.88 | 0.60-5.90 | 2 | 0.59 | 0.13-2.79 | |
| without | ≥50 | 1 | 0.96 0.12-1.42 | 5 | 1.01 | 0.28-3.59 | 2 | 0.59 | 0.12-2.92 | 0 | ||
| "highly exposed"" | ||||||||||||
| 2-year latency period | ||||||||||||
| All | 0 -< 10 | 8 | Reference | 4 | Reference | 5 | Reference | 7 | Reference | |||
| patients | 10 -< 50 | 0 | 2 | 0.71 | 0.13-3.93 | 6 | 1.95 | 0.58-6.48 | 2 | 0.60 | 0.12-2.88 | |
| ≥50 | 0 | 4 | 1.35 | 0.34-5.44 | 2 | 0.60 | 0.12-3.19 | 0 | ||||
| All | 0 -< 10 | 8 | Reference | 4 | Reference | 5 | Reference | 7 | Reference | |||
| patients | 10 -< 50 | 0 | 1 | 0.36 | 0.04-3.21 | 5 | 1.63 | 0.46-5.72 | 2 | 0.62 | 0.13-2.97 | |
| without | ≥50 | 0 | 4 | 1.38 | 0.34-5.57 | 2 | 0.61 | 0.12-3.19 | 0 | |||
| "highly exposed"" | ||||||||||||
ª IRR: incidence rate ratio, CI: 95% Wald confidence interval.
b Effective dose for blastomas, sarcomas and other solid tumors, organ dose for CNS tumors (brain dose).
c See methods section for a definition.
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46
1 2 3 4 5
| Study Tags and labels of ICD-10ª blocks and diseases | ICD-10 Codes |
|---|---|
| Patients tagged as having "high mortality risk" when seen at DvHCH radiology department | |
| Tuberculosis, Meningococcal infection, Streptococcal septicaemia, Other septicaemia | A15-A19, A39-A41 |
| Human immunodeficiency virus (HIV) disease | B20-B24 |
| Other coagulation defects | D68 |
| Cystic fibrosis, Other metabolic disorders | E84, E88.0 |
| Mental and behavioural disorders due to psychoactive substance use | F10-F19 |
| Inflammatory diseases of the central nervous system, | G00-G09, G10-G13, G80- |
| Systemic atrophies primarily affecting the central | G83, G90-G99 |
| nervous system, Cerebral palsy and other paralytic syndromes, Other disorders of the nervous system | |
| Acute rheumatic fever, Chronic rheumatic heart | 100-102, 105-109, 126, 130, 138, |
| diseases, Pulmonary embolism, Endocarditis, Cardiomyopathy, Cerebrovascular diseases, Atherosclerosis, Aortic aneurysm and dissection, Other aneurysm | 143, 160-169, 171-172 |
| Acute epiglottitis, Chronic laryngitis and laryngotracheitis | J05.1, J37 |
| Appendicitis, Diverticular disease of intestine, Fissure and fistula of anal and rectal regions, Other diseases of anus and rectum, Peritonitis, Alcoholic hepatic failure, Toxic liver disease with hepatic necrosis, Hepatic failure, Fibrosis and cirrhosis of liver | K35-K36, K57, K60, K62, |
| K65, K70.4, K71.1, K72, K74 | |
| Mucocutaneous lymph node syndrome (Kawasaki) | M30.3 |
| Renal failure | N17-N19 |
| Chronic respiratory disease originating in the perinatal period, Necrotizing enterocolitis of fetus and newborn, Other disturbances of cerebral status of newborn | P27, P77, P91 |
| Congenital malformations, deformations and | Q00-Q07, Q10-Q18, Q20- |
| chromosomal abnormalities (except Down's syndrome | Q28, Q30-Q34, Q35-Q37, |
| and few other syndroms) | Q50-Q56, Q60-Q64, Q65- |
| Q79, Q80-Q89 except Q18.1, | |
| Q52.8, Q65.8, Q66, Q67.6, | |
| Q67.7, Q69, Q70 | |
| Injury involving multiple body regions, Poisoning | T00-T07, T36-T50 |
| Patients tagged as having "elevated incidence risk" | |
| Agranulocytosis, Immunodeficiency | D70, D80-D83 |
| Crohn's disease, Colitis ulcerosa | K50-K51 |
| Down's syndrome, other chromosomal anomalies | Q90-Q99 |
ª International Coding of Diseases, 10th revision.
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
| 5-year latency period | ||||
|---|---|---|---|---|
| 0ª | Eª | SIRª | 95% CIª | |
| CNS tumors | 7 | 10.46 | 0.67 | 0.27-1.38 |
| Blastomas | 7 | 1.96 | 3.57 | 1.44-7.36 |
| Sarcomas | 7 | 7.31 | 0.96 | 0.38-1.97 |
| Other solid tumors | 7 | 2.77 | 2.53 | 1.02-5.20 |
ª O: observed cases, E: expected cases, SIR: standardized incidence ratio, CI; confidence interval.
Accepted Manuscript
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2 3
| Group | Dose categoryb (uSv) | CNS tumors (III) | Blastomas (IV-VII) Man | Sarcomas (VIII-IX) | Other solid tumors (X- XII) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Case IRRª S | 95% CIª | Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | Cases | IRRª | 95% CIª | ||
| 5-year latency period | ||||||||||||
| All | 0 -< 10 | 7 Reference | 2 | Reference | 3 | Reference | 5 | Reference | ||||
| patients | 10 -< 50 | 0 | 2 | 2.47 | 0.22-27.35 | 3 | 1.48 | 0.30-7.36 | 2 | 0.63 | 0.12-3.26 | |
| ≥50 | 0 | 4 | 4.73 | 0.53-42.43 | 1 | 0.47 | 0.05-4.55 | 0 | ||||
| All | 0 -< 10 | 7 | Reference | 1 | Reference | 3 | Reference | 5 | Reference | |||
| patients | 10 -< 50 | 0 | 1 | 1.26 | 0.08-20.25 | 2 | 1.02 | 0.17-6.11 | 2 | 0.65 | 0.13-3.35 | |
| without | ≥50 | 0 | 4 | 4.94 | 0.55-44.33 | 1 | 0.49 | 0.05-4.74 | 0 | |||
| "highly exposed"" | Accepted | |||||||||||
ª IRR: incidence rate ratio, CI: 95% Wald confidence interval.
b Effective dose for blastomas, sarcomas and other solid tumors, organ dose for CNS tumors (brain dose).
” See methods section for a definition.
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46