Revised Manuscript: Ref.: Ms. No. PNEP-D-13-00561
Subsequent cancer risk of children receiving post voiding cystourethrography: A
nationwide population-based retrospective cohort study
Running title: VCUG and cancer risk
Word counts: 2210
Yen-Hsiu Liao, MD1,2, Cheng-Li Lin, MSc3,4, Wei Chang Ching, MD5, Po-Pang Tsai,
MD1,2, Wu-Chung Shen, MD1,6, Fung-Chang Sung, PhD, MPH3,4 , Tsai-Chung Li,
PhD4, Chia-Hung Kao, MD7,8*
1
Department of Radiology, China Medical University Hospital, Taichung, Taiwan;
2
School of Medicine, College of Medicine, China Medical University, Taichung,
Taiwan; 3Management Office for Health Data, China Medical University Hospital,
Taichung, Taiwan; 4Department of Public Health, China Medical University,
Taichung, Taiwan; 5Department of Pediatrics, China Medical University Hospital,
China Medical University, Taichung, Taiwan; 6Department of Biomedical Imaging
and Radiological Science, College of Health Care, China Medical University,
1
Taichung, Taiwan; 7Graduate Institute of Clinical Medical Science, School of
Medicine, College of Medicine; 8Department of Nuclear Medicine and PET Center
China Medical University Hospital, Taichung, Taiwan
Corresponding Author: Chia-Hung Kao, MD, Graduate Institute of Clinical
Medicine Science and School of Medicine, College of Medicine, China Medical
University, No. 2, Yuh-Der Road, Taichung 404, Taiwan. Tel.: +886 4
22052121x7412; Fax.: +886 4 22336174. E-mail: d10040@mail.cmuh.org.tw
Author Contributions:
Study concept and design: Yen-Hsiu Liao, Chia-Hung Kao.
Acquisition of data: Wei Chang Ching, Po-Pang Tsai, Wu-Chung Shen, Fung-Chang
Sung, Tsai-Chung Li.
Analysis and interpretation of data: Yen-Hsiu Liao, Cheng-Li Lin, Wei Chang Ching,
Chia-Hung Kao.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: Yen-Hsiu Liao,
Chia-Hung Kao.
Statistical analysis: Cheng-Li Lin.
2
Obtained funding: Fung-Chang Sung, Tsai-Chung Li, Chia-Hung Kao
Administrative, technical, or material support: Fung-Chang Sung, Tsai-Chung Li.
Study supervision: Chia-Hung Kao
Conflict of Interest Disclosures: All authors have completed and submitted the
ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Potential Conflicts of Interest and none were reported.
Funding/Support: The study was supported in part by Taiwan Department of Health
Clinical Trial and Research Center and for Excellence (DOH102-TD-B-111-004),
Taiwan Department of Health Cancer Research Center for Excellence
(DOH102-TD-C-111-005), Bureau of Health Promotion, Department of Health,
R.O.C. (Taiwan) (DOH99-HP-1205 ), and International Research-Intensive Centers of
Excellence in Taiwan (I-RiCE) (NSC101-2911-I-002-303).
Role of the Sponsor: Taiwan Department of Health had no role in the design and
conduct of the study; in the collection, analysis, and interpretation of the data; or in
the preparation, review, or approval of the manuscript.
3
Disclaimer: The interpretations and conclusions contained in this article do not
represent those of the Bureau of National Health Insurance, Department of Health, or
the NHRI. National Health Insurance Research Database, Taiwan.
http://www.nhri.org.tw/nhird/en/index.htm.
4
ABSTRACT
Background: To estimate the subsequent cancer risk of children receiving post
voiding cystourethrography (VCUG), a nationwide population-based retrospective
cohort study with the data from the Taiwan National Health Insurance Research
Database (NHIRD) were used for the analysis.
Methods: In the VCUG cohort, 31 908 participants younger than 18 years of age,
who underwent VCUG between 1997 and 2008, were identified from the NHIRD. A
comparison cohort, the non-VCUG cohort, was randomly selected among children
without VCUG examination histories during 1997–2008, frequency matched for age
(every 5 y), sex, geographic region area, parents’ occupation, and index year based on
a 1:4 ratio. Cox’s proportional hazard regression analysis was conducted to estimate
the subsequent cancer risk of children receiving VCUG.
Results: The overall cancer risk of the VCUG cohort is 1.92-fold (95% CI =
1.34–2.74) higher than that of the non-VCUG cohort with statistical significance. The
genital cancer and urinary system cancer risks of the VCUG cohort are respectively
6.19-fold (95% CI = 1.37–28.0) and 5.8-fold (95% CI = 1.54–21.9) higher than those
of the non-VCUG cohort with statistical significance. The hazard ratios are higher in
genital cancer, urinary system cancer (the major radiation exposure area), and cancer
5
of the abdomen, except for the genitourinary system (the minor radiation exposure
area), in sequence.
Conclusions: Pediatric VCUG is associated with increased subsequent cancer risk,
especially in the genitourinary system.
Keywords: voiding cystourethrography; radiation-induced cancer; cohort study.
6
Introduction
Vesicoureteral reflux (VUR), the most common heritable disorder of the
genitourinary tract, is the retrograde flow of urine from the bladder into the ureter,
renal pelvis, and/or kidney [1]. VUR is common in children and constitutes a critical
health impact in children, because the complications of VUR-associated
pyelonephritis are highly relevant. A total of 6.2% of low-grade VUR (Grades I–III)
and 47.9% high-grade VUR (Grades IV and V) have renal damage [2]. The standard
diagnostic test for VUR is a relatively invasive voiding cystourethrogram (VCUG)
[3-4]. VCUG is the most commonly performed (up to 50%) fluoroscopic examination
on children [5]. However, in addition to possible iatrogenic traumatic injury, the other
critical complication is the radiation effect on children, who are more sensitive to
radiation injury than adults are. Furthermore, children typically have a longer
remaining lifespan than adults do. The effects of radiation injury caused by radiation
dose accumulation are, therefore, more critical in children [6-8].
Based on the fluoroscopic technical development and radiation protection ALA
RA concept (as low as reasonably achievable), the effective doses of pediatric VCUG
were 0.1 to 0.5 mSv in the most recent phantom study [9]. According to the entire
population cancer risk estimation provided by the BEIR VII [10], this effective dose
might be associated with a lifetime risk of fatal cancer induction of 0.4–3/100 000 in
7
the general population. The lifetime cancer mortality risk caused by radiation injury is
estimated at 3-fold in children compared with adults [11]. Therefore one would expect
the lifetime risk of fatal cancer induction post VCUG to be higher than 0.4–3/100 000
in children. To estimate the cancer hazard ratio of children post VCUG in comparison
to children not receiving VCUG, we conducted a nationwide large population-based
cohort study.
8
Methods
Data sources
The nationwide cohort study was based on patient data obtained from the
reimbursement claims of the universal National Health Insurance Research Database
(NHIRD), managed by the Taiwan National Health Research Institutes (NHRI) [12].
The claims data used in this study represented information for half of all children,
randomly selected from all insured population aged ≤18 years in Taiwan. Claims data
contain basic demographic information on insured children (eg, sex, birth date, and
residential area) and medical care received for ambulatory and inpatient visits. Cancer
diagnosis was defined according to the Registry for Catastrophic Illness Patient
Database (RCIPD), which is a separate subpart of the NHIRD. The diagnoses in the
database were coded using the International Classification of Diseases 9th Revision of
Clinical Modification (ICD-9-CM). The study was approved by the NHRI.
Study participants
We identified 31 908 child patients who had received an initial voiding
cystourethrogram (VCUG) examination among participants aged 1 to 18 years during
9
1997–2008. The index date was that of VCUG examination. We excluded child with
histories of malignant cancer before the index date or with missing information on sex
and age. A comparison cohort was randomly selected among children without VCUG
examination histories during 1997–2008, frequency matched for age (every 5 y), sex,
geographic region, parents’ occupation, and index year based on a 1:4 ratio.
Outcome definition
Every child with a cancer diagnosis (ICD-9 codes 140–194, and 200–208) was
identified by the RCIPD. Each study participant was followed-up to evaluate the
occurrence of cancer until December 31, 2010, or censored because of death,
withdrawal from NHI, or loss to follow-up. The categories of cancer classification
were considered, including cancer in the abdomen except the genitourinary system
(ICD-9 codes 151–159, 180, 182, 184, and 187), cancer of the neuroendocrine system
including the eyes (ICD-9 code 190-192, 194), cancer out of the abdomen (ICD-codes
160-165, 170-176), genital cancer (ICD-9 codes 183, 186), urinary system cancer
(ICD-9 codes 188-189), hematologic system cancer (ICD-9 codes 200-208), and other
cancer types.
10
Statistical analysis
Distributions of categorical socio-demographic factors, including age (≤ 8, 9–18
y), residential area (Northern, Central, Southern, and Eastern Taiwan), parental
occupation (white collar, blue collar, or other), were compared between cohorts with
and without VCUG examination. We used the chi-square test for categorical variables
and the t-test for continuous variables. We calculated the incidence rate of cancer in
both cohorts. The Cox proportional hazards regression analysis was used to estimate
the hazard ratios (HRs) and 95% confidence intervals (CIs) of cancer and site-specific
cancers for the cohort with VCUG examination relative to the cohort without VCUG
examination. Both crude HRs and multivariable adjusted HRs were estimated. We
used the Kaplan-Meier method to compare the probability of cancer-free events and
used the log-rank test to examine the significance of the difference between the two
cohorts. SAS version 9.2 (SAS Institute, Cary, NC, USA) was used for data analyses.
All significance levels were set at a 2-tailed P <.05. Kaplan-Meier curves were plotted
using R (version 2.14.1; R Development Core Team, Vienna, Austria).
11
Results
In our study, the children with VCUG examination were more likely to be boys
(58.7%), ≤ 5 years of age (85.8%) (Table1). Child patients with VCUG examinations
were more likely residing in the northern (53.4%) region, and the parental occupations
were more likely white-collar (65.7%). Table 2 presents the incidence rates in both
cohorts and the VCUG-to-non-VCUG HR of cancer according to socio-demographic
factors. In total, we observed 151 cases of cancer (99 in the non-VCUG cohort, 52 in
the VCUG cohort) among 707 762 person-years, with an incidence rate of 14.0 per
100 000 person-years for the non-VCUG cohort and 29.3 per 100 000 person-years
for the VCUG cohort (Crude HR = 2.10, 95% CI = 1.50–2.93). The multivariate Cox
proportional hazard regression analysis revealed that the risk of developing cancer
among the VCUG cohort was 1.92-fold higher than that of the non-VCUG cohort
(adjusted HR = 1.92, 95% CI = 1.34–2.74). The highest age-specific and sex-specific
HR were observed in patients ≤ 8 years (adjusted HR = 1.91, 95% CI = 1.32–2.74)
and in the boy patients (adjusted HR = 1.82, 95% CI = 1.16–2.86). Region-specific
analysis showed that the highest incidence rate for those living in the eastern area with
a VCUG examination was 49.4 per 100 000 person-years in the VCUG cohort.
Moreover, Patients living in the eastern area had the highest risk of developing cancer
with statistical significance (adjusted HR = 2.16, 95% CI = 1.28–3.62). Patients with
12
parents in white-collar occupations had the highest risk of developing cancer with
statistical significance (adjusted HR = 2.42, 95% CI = 1.52–3.95). The specific
analyses of different cancer types between the VCUG cohort and the non-VCUG
cohort are shown in Table 3. Compared to the non-VCUG cohort, the adjusted HR of
developing cancer in the genitals, urinary system, and hematologic system were
6.19-fold (95% CI = 1.37–28.0), 5.80-fold (95% CI = 1.54–21.9), and 1.82-fold (95%
CI = 1.05–3.13) for the VCUG cohort, respectively. Figures 1(a)–1(d) show that the
VCUG cohort had a significantly higher cumulative proportion of cancer (P < .0001)
(Figure 1a), genital cancer (P = .015) (Figure 1b), urinary system cancer (P = .0003)
(Figure 1c), and hematologic system cancer (P = .007) (Figure 1d) compared with the
non-VCUG cohort.
13
Discussion
Based on thorough research, this is the first nationwide large population-based
cohort study to investigate the cancer hazard ratio of children receiving post VCUG in
comparison with children not receiving VCUG. The overall cancer risk of the VCUG
cohort is 1.92-fold (95% CI = 1.34–2.74) higher than that of the non-VCUG cohort
with statistical significance. With stratification according to age, sex, resident
geographic area, and parents’ occupation, all cancer risks of the VCUG cohort are
higher than those of the non-VCUG cohort by 1.29–2.93 fold. Among them, the
hazard rations of ages less than 8 years old, male sex, northern residential area, and
white collar occupation are statistically significant. With stratification according to
cancer location, the ovary cancer and testes cancer risks of the VCUG cohort are
6.19-fold (95% CI = 1.37–28.0) higher than those of the non-VCUG cohort with
statistical significance. The urinary system cancer risk (including kidney, ureter,
urinary bladder, and urethra) of the VCUG cohort is 5.8-fold (95% CI = 1.54–21.9)
higher than that of the non-VCUG cohort with statistical significance. The risks of
other cancers of the abdomen except the genitourinary system and the hematologic
system (eg, leukemia and lymphoma) are also higher (1.82–2.98-fold) in the VCUG
cohort than in the non-VCUG cohort with statistical significance in the hematologic
cancer group. The hazard ratios are higher for genital cancer, urinary system cancer
14
(the major radiation exposure area), and cancer of the abdomen except the
genitourinary system (the minor radiation exposure area) in order. The risk of
non-abdominal cancers except hematologic cancer and neuroendocrine cancer (out of
the radiation exposure area) are lower in the VCUG cohort than in the non-VCUG
cohort.
A higher accumulation of a radiation dose leads to a greater probability of
carcinogenesis and genetic mutations. In the past decade, clinicians and pediatric
radiologists have worked hard to reduce radiation dose accumulation in children by
using VCUG. The American Urological Association and the American Academy of
Pediatrics have recommended that a VCUG study should not be performed routinely
after the first febrile urinary tract infection in children. A VCUG study is indicated
when a renal bladder ultrasound reveals hydronephrosis, scarring, or other findings
suggests high-grade VUR, obstructive uropathy, other atypical, or complex clinical
circumstances, in their newest guidelines [13-14]. Pediatric radiologists and
radiologic technicians have studied methods for reducing fluoroscopic radiation doses
by using urine sensor devices, pulse fluoroscopy, last image hold, automatic
anatomical programming, and less fluoroscopy time by an experienced performer
[15-17]. Other image modalities, including voiding urosonography with ultrasound
contrast media and magnetic resonance voiding cystourethrography, have also been
15
studied for free radiation. However, the traditional fluoroscopic VCUG is still the gold
standard diagnostic test for VUR because of its more accurate determination of the
grade of VUR and higher-quality assessment of the bladder and urethral configuration
[18]. The limitations of voiding urosonography are its extremely operator-dependent
sensitivity from 63% to 100% [19-20] and the insufficient delineation of the entire
urinary system. The limitations of magnetic resonance VCUG are the relatively low
successful completion rate of 76.7%, sedation among children, cost, and rarity [21].
The limitation of this study is the difficulty of defining the actual exposure of
X-ray equipment because such equipment is designed differently. Exposure frequency
may also be underestimated because the X-ray database was collected by contracted
Taiwan National Health Insurance (TNHI) practitioners and excluded non-TNHI data
(including self-paying patents). In addition, X-ray data received prior to 1996 are
unavailable in the National Health Research Institute data sets. Therefore,
misclassification of the X-ray exposure status is possible, and some patients were thus
included in the non-exposure group. However, the large sample size of the VCUG
cohort from nationwide population-based data sets strengthens the statistical power of
our study on the associations between pediatric VCUG and subsequent cancer risks.
Furthermore, the nearly comprehensive coverage of the nationwide health insurance
enables reducing the likelihood of loss to follow-up.
16
The other limitation of this study is the data regarding the interval from radiation
exposure to subsequent abdominal/GU cancers could not be obtained exactly due to
the record factor. Beside VUR, the children who are suspected of congenital
genitourinary anomaly are prone to receive VCUG. And the children with congenital
genitourinary anomaly have more probability of Wilms’ tumor. However, due to the
small number of children with congenital genitourinary anomaly and the lack of
follow-up in the literature, no conclusions could be made above the carcinoma
development in children with congenital genitourinary anomaly [22-23].
In conclusion, we found that the overall cancer risk of the VCUG cohort is
1.92-fold (95% CI = 1.34–2.74) higher than that of the non-VCUG cohort with
statistical significance. A higher accumulation of a radiation dose leads to a greater
probability of carcinogenesis and genetic mutations. Because VCUG is still the
standard diagnostic test for VUR, clinicians and pediatric radiologists should always
follow the ALA RA radiation protection concept (as low as reasonably achievable)
from making decisions to performing procedures, as preventing unnecessary radiation
examination and lowering the examination radiation dose as reasonably achievable in
the necessary examination, to reduce radiation accumulation dose, and, therefore,
reduce the cancer risk in children.
17
References
1. Hollowell JG, Greenfield SP. Screening siblings for vesicoureteral reflux. J Urol
2002; 168:2138–2141.
2. Skoog SJ, Peters CA, Arant BS Jr, et al. Pediatric vesicoureteral reflux guidelines
panel summary report: clinical practice guidelines for screening siblings of children
with vesicoureteral reflux and neonates/infants with prenatal hydronephrosis. J Urol
2010; 184:1145–1151.
3. Williams G, Fletcher JT, Alexander SI, Craig JC. Vesicoureteral reflux. J Am Soc
Nephrol 2008; 19:847–862.
4. Hannula A, Venhola M, Perhomaa M, et al. Imaging the urinary tract in children
with urinary tract infection. Acta Paediatr 2011; 100:e253–e259.
5. Sulieman A, Theodorou K, Vlychou M, et al. Radiation dose measurement and risk
estimation for paediatric patients undergoing micturating cystourethrography. Br J
Radiol 2007; 80:731–737.
6. Claus EB, Calvocoressi L, Bondy ML, Schildkraut JM, Wiemels JL, Wrensch M.
Dental x-rays and risk of meningioma. Cancer 2012; 118:4530–4537.
Claus EB, Calvocoressi L, Bondy ML, Schildkraut JM, Wiemels JL, Wrensch M.
7. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure from CT scans in
childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort
18
study. Lancet 2012;380:499–505.
8. Lin MC, Lee CF, Lin CL, et al. Dental diagnostic X-ray exposure and risk of
benign and malignant brain tumors. Ann Oncol 2013;24:1675–1679.
9. Lee R, Thomas KE, Connolly BL, et al. Effective dose estimation
for pediatric voiding cystourethrography using an anthropomorphic phantom set and
metal oxide semiconductor field-effect transistor (MOSFET) technology. Pediatr
Radiol 2009; 39:608–615.
10. National Research Council. Health risks of exposure to low levels of ionizing
radiation: BEIR VII. National Academies Press, Washington DC, 2006.
11. ICRP. 1990 Recommendations of the International Commission on Radiological
Protection. Publication 60. Annals of the ICRP 21(1-3). Pergamon, Oxford, 1991.
12. Cheng TM. Taiwan’s National Health Insurance system: high value for the dollar.
In Okma, K.G.H. and Crivelli, L. ed. Six Countries, Six Reform Models: The Health
Reform Experience of Israel, the Netherlands, New Zealand, Singapore, Switzerland
and Taiwan. New Jersey: World Scientific, 2009, pp.71-204.
13. Peters CA, Skoog SJ, Arant BS, et al. Summary of the AUA guideline on
management of primary vesicoureteral reflux in children. J Urol 2010;
184:1134–1144.
14. Subcommittee on Urinary Tract Infection, Steering Committee on Quality
19
Improvement and Management. Urinary tract infection: Clinical Practice guideline for
the diagnosis and management of the initial UTI in febrile infants and children 2 to 24
months. Pediatrics 2011; 128:595–610.
15. Ward VL. Patient dose reduction during voiding cystourethrography. Pediatr
Radiol 2006; 36:168–172.
16. Ward VL, Strauss KJ, Barnewolt CE, et al. Pediatric radiation exposure and
effective dose reduction during voiding cystourethrography. Radiology 2008;
249:1002–1009.
17. Brook OR, Brook A, Soudack M. Urine sensor device for fluoroscopy time
reduction in pediatric voiding cystourethrography. Eur J Radiol 2011; 78:394–397.
18. Tekgül S, Riedmiller H, Hoebeke P, et al. EAU guidelines on vesicoureteral reflux
in children. Eur Urol 2012; 62:534–542.
19. Darge K. Voiding urosonography with US contrast agent for the diagnosis of
vesicoureteric reflux in children: an update. Pediatr Radiol 2010;40:956–962.
20. Galloy MA, Guillemin F, Couture A, et al. Voiding ultrasonography: evaluation of
the detection of vesicoureteral reflux based on the review of digital ultrasound clips.
Ultraschall Med 2008; 29:53–59.
21. Johnin K, Takazakura R, Furukawa A, et al. Magnetic resonance voiding
cystourethrography (MRVCUG): A potential alternative to standard VCUG.
20
J Magn Reson Imaging. 2013 Feb 15. doi: 10.1002/jmri.24052. [Epub ahead of print]
22. Pendergrass TW. Congenital anomalies in children with Wilms' tumor: a new
survey. Cancer 1976;37:403–408.
23. Woods MS, Sheppard RG, Hardman DA, Woods HJ. Congenital genitourinary
anomalies. Is there a predilection for multiple primary malignant neoplasms? Cancer.
1992;69:546–549.
21
Figure Legend:
Figure 1 The Kaplan-Meier curves of the all cancer-free rate (a), genital cancer (b),
GU cancer (c) and the hematologic cancer-free rate (d) in patients with or without
receiving VCUG examination
22