Radiation hazards to patients and medical workers exposed to radiological examinations have raised worldwide concerns in recent decades (1–3). The ICRP 2015 Proceedings indicate that 32.7 million global diagnostic nuclear medicine (NM) examinations are performed annually, with an increase of 0.2 million examinations per year since 1991 (4). According to the National Council on Radiation Protection and Measurements (NCRP) 160 report, NM procedures have increased from 6.3 million in 1984 to 18 million in 2006, with ~1% of these procedures performed on children in the United States (5). NM imaging provides essential physiological and pathological information in oncology, urology, and orthopedics, and such imaging is particularly valuable in children and infants, in whom a rapid and accurate diagnosis is crucial for developmental health and reducing disease progression. Children have a higher risk for adverse effects from radiation exposure than adults, and the subsequent lifelong estimated cancer risks due to repeated radiological examination should not be ignored (6, 7). NM procedures are traditionally thought to be safe and non-invasive without serious complications. For example, ^99mTc-DMSA renal scintigraphy can be used acutely to confirm the presence of acute pyelonephritis or, 4–6 months later, the sequelae of renal cortical scarring in pediatric patients with recurrent urinary tract infections. The estimated effective dose administered for a common ^99mTc-DMSA scan is low, varying between 0.6 and 0.8 millisieverts (mSv). Although the majority of NM exams performed on children require a low effective dose of <1 mSv, radiation exposure and cancer risk from repeated scans in the acute stage and disease follow-up should never be overlooked (8). Some NM examinations, especially in oncology and orthopedics, can result in an intermediate radiation dose, between 3 and 6 mSv, which is close to the radiation dose received from one modern CT scan (9, 10). Several studies have revealed evidence of higher cancer risk after repeated CT scan exposure in pediatric patients (11–15). It is conceivable that the cancer risk associated with NM procedures in the pediatric population is a topic that warrants further assessment. Our encrypted identification and medical records of participants were representatively extracted from the NHIRD, which has the advantage of being one of the largest national databases in the world. This population-based study was designed to investigate the possible association between pediatric neoplasms and diagnostic NM procedures.

This current study shows that pediatric patients who are exposed to NM examinations have a significantly higher risk of developing neoplasms than non-exposed subjects. The incidence rates of any malignant tumor and benign brain and central nervous neoplasm in the NM scanning cohort were significantly higher than those of patients without scanning (16.9 vs. 1.54; adjusted HR, 10.9; 95% CI, 6.53–18.2). Subsequent analysis revealed the trend that with an increasing number of NM scans, the risk of pediatric neoplasms was aggregated (increase in adjusted HR from 7.50 to 56.4 with the increase in frequency from 1–2 to ≥5).

The most significant consequence of low-level radiation exposure in humans is cancer. The carcinogenic risk of radiation may be two to three times higher in children than in adults, although this risk varies with organ and tissue type (8). Cancer induction resulting from radiation exposure occurs in a stochastic manner: there is no threshold point, and the risk increases proportionally with the dose (17). The vulnerability of the pediatric population to ionizing radiation is due not only to the enhanced radiosensitivity of their tissues but also to a longer time period for the manifestation of stochastic radiation effects. First, our study results revealed that the incidence rate of pediatric neoplasms was significantly higher in the NM exposure group from the beginning of the follow-up period, which is consistent with the lack of a stochastic effect threshold. Furthermore, a higher risk of cancer is significantly associated with a higher frequency of NM exams, which indicates higher radiation dose exposure. This proportional probability of cancer occurrence is also compatible with the dose-dependent stochastic effect (Table 2; Figure 2). The abovementioned two results demonstrate the traditionally accepted linear and no-threshold model of the stochastic radiation effect.

Hereditary deformity attributed to low-dose radiation is the other example of a stochastic effect. Earlier studies linked prenatal medical radiation exposure to pediatric cancer in offspring, but there was no solid evidence for these conclusions (18, 19). Recent articles reviewed animal studies or applied different risk estimation models to discuss low-dose radiation exposure in nuclear medicine and the associated risk of cancer (20, 21). A PubMed-based literature search revealed a lack of large-scale investigations into childhood NM radiation exposure and subsequent cancer risk. One of the possible reasons is difficulties in pediatric NM procedure dose estimation. NM radiation dosimetry is challenging due to diversities in the uptake, retention, and clearance of variable radiopharmaceuticals. Furthermore, individual administered doses in children also differ depending on the patient's body mass, type of examination, available SPECT model, examination time, and patient cooperation (22, 23). Nevertheless, updated nuclear medicine drug decay half-lives are now available in ICRP Publication 107 (24). Based on these contemporary radiopharmaceutical data and the new biokinetic model, more accurate absorbed dose estimations are available for patients examined with NM procedures (25–27). With advances in absorbed dose estimation, we expect more quantitative knowledge about the risk of cancer associated with pediatric NM procedures in the future. At present, parents of pediatric patients are always anxious about the lifelong cancer risk associated with NM procedures. According to our study results, cancer risk associated with pediatric NM examinations should never be ignored. As a result of this issue, we advise physicians to take a more comprehensive approach and to consider the benefits of nuclear medicine as well as the potential risk associated with radiation exposure when deciding the best option for pediatric patients.

The HWDC datasets do not provide information about the clinical purpose of NM examinations. Thus, a screening effect and selection bias in the NM exposure cohort may have been present in this study. Pediatric patients who underwent NM procedures might receive additional radiation exposure, such as through routine plain films or other special examinations. We excluded high radiation dose procedures such as CT scan, radiotherapy, PET scan, and therapeutic NM procedures in both the exposure and unexposed groups, and this confounding factor might not be strong enough to affect the significance of our result. In addition, the NHIRD (National Health Insurance Research Database) cannot provide radiopharmaceuticals' dosage from each NM examination, the lack of the accurate data about radiopharmaceuticals decay (related to the accurate absorbed radiation and exposure) is our major study limitation.

Therefore, due to the lack of data about the actual absorbed radiation dose at each NM examination, our results indicating higher pediatric cancer risk associated with NM exposure should be interpreted with caution. Besides above mentioned limitation, prenatal information and family history of our participants were not included in the HWDC datasets, and hereditary or genetic effects of developing pediatric neoplasms could not be assessed.

In conclusion, this study benefits from a long follow-up period (2000–2018) and a large sample size (35,292 subjects with NM exposure) and shows that low-dose radiation exposure from NM examinations is significantly associated with a higher risk of cancer and benign central nervous tumors in pediatric patients. Nuclear medicine practitioners, including physicians and technologists, should show more initiative to protect patients from radiation and should be prudent in optimizing the radiation dose in pediatric NM procedures.

The dataset used in this study is held by the Taiwan Ministry of Health and Welfare (MOHW). The Ministry of Health and Welfare must approve our application to access this data. Any researcher interested in accessing this dataset can submit an application form to the Ministry of Health and Welfare requesting access. Please contact the staff of MOHW (Email: stcarolwu@mohw.gov.tw) for further assistance. All relevant data are within the paper. Requests to access the datasets should be directed to stcarolwu@mohw.gov.tw.

The studies involving human participants were reviewed and approved by the Central Regional Research Ethics Committee, China Medical University, Taichung, Taiwan (CMUH109–109-REC2–031). Written informed consent from the participants' legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

M-KY and C-HK: conception and design. C-HK: administrative support. All authors contributed significantly, agreement with the content of the manuscript, collection and assembly of data, data analysis and interpretation, manuscript writing, and final approval of manuscript.

This study was supported in part by Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW110-TDU-B-212-124004), China Medical University Hospital (DMR-109-231, DMR-110-089, DMR-111-105), Ministry of Science and Technology (MOST 110-2321-B-039-003). The funders had no role in the study design, data collection and analysis, the decision to publish, or preparation of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.