Original Article
Risk of Cardiovascular Disease Among Nordic Childhood
Cancer Survivors With Diabetes Mellitus: A Report From
Adult Life After Childhood Cancer in Scandinavia
Jeanette F. Winther, MD, DMSc1,2; Smita Bhatia, MD, MPH3; Luise Cederkvist, PhD1; Thorgerdur Gudmundsdottir, MD, PhD1;
Laura Madanat-Harjuoja, MD, PhD4; Laufey Tryggvadottir, MSc 5,6; Finn Wesenberg, MD, PhD7,8,9;
Henrik Hasle, MD, PhD10; and Anna Sällfors Holmqvist, MD, PhD11,12; for the ALiCCS Study Group
BACKGROUND: Childhood cancer survivors have an increased risk of cardiovascular disease (CVD) and diabetes mellitus. Because
diabetes is a potentially modifiable risk factor for CVD in the general population, it is important to understand how diabetes affects
the risk of CVD among childhood cancer survivors. METHODS: This study examined the risk of CVD among survivors with diabetes
and 142,742 population comparison subjects. From the national cancer registries of the 5 Nordic countries, 29,324 one-year survivors
of cancer diagnosed before the age of 20 years between 1968 and 2008 were identified. Study subjects were linked to the national
hospital registers. The cumulative incidence of CVD was determined with competing risk methods. A Cox proportional hazards
model was used to estimate the effects of diabetes and cancer on the hazard of CVD. The interaction between diabetes and cancer
was analyzed. RESULTS: Diabetes was diagnosed in 324 of the 29,324 one-year survivors, and CVD was diagnosed in 2108. The
hazard of diabetes was 1.7 times higher among survivors than comparison subjects (hazard ratio [HR], 1.7; 95% confidence interval
[CI], 1.5-1.9), whereas the HR of CVD was 3.6 (95% CI, 3.3-3.8) 1 to 15 years after the cancer diagnosis and 1.9 (95% CI, 1.8-2.0) after
more than 15 years. Individuals with diabetes had a 2.4 times higher hazard of CVD (95% CI, 2.1-2.8) among both survivors and comparison subjects in comparison with individuals without diabetes. CONCLUSIONS: Childhood cancer survivors with diabetes have a
markedly increased risk of CVD in comparison with survivors without diabetes. However, diabetes does not increase the risk of CVD
more in survivors than the general population. Cancer 2018;124:4393-4400. © 2018 American Cancer Society.
KEYWORDS: cardiovascular disease, cerebrovascular disease, childhood cancer, diabetes mellitus, survivorship.
INTRODUCTION
The remarkable improvement in the treatment of childhood cancer during the past 5 decades has resulted in a rapidly
growing population of long-term survivors.1 However, these advances in treatment and survival come at a price, and
many survivors face an increased risk for a broad range of treatment-induced sequelae, most of which become clinically
apparent many years after the children have been cured.2,3 It is well established that childhood cancer survivors have an
increased risk of cardiovascular morbidity and mortality, which is associated with exposure to anthracycline therapy and
chest-directed radiotherapy.4-8 Previous research has also shown that childhood cancer survivors have an increased risk
for diabetes mellitus.9-13 Diabetes mellitus is, in turn, a well-known risk factor for cardiovascular disease (CVD) in the
general population.14-16 Yet, within the survivor cohort of the Childhood Cancer Survivor Study (CCSS), diabetes was
not found to potentiate cancer therapy–associated risk for CVD except for the risk of heart failure among those treated
with chest-directed irradiation.17
In the general population, several studies have shown that early diagnosis and improved treatment of diabetes
decrease the risk of complications of diabetes, including CVD.16,18,19 Therefore, it is important to fully understand
how diabetes, a potentially modifiable risk factor, affects the risk of CVD among childhood cancer survivors and
whether the impact of diabetes on the risk of CVD in survivors differs from that in the general population. Using the
unique resources of Adult Life After Childhood Cancer in Scandinavia (ALiCCS), a large population-based cohort of
Nordic childhood cancer survivors, we address these questions by examining the risk of CVD overall and the risk for
Corresponding author: Anna Sällfors Holmqvist, MD, PhD, Department of Clinical Sciences, Lund University, Lasarettsgatan 40, SE-221 85, Lund, Sweden;
anna.sallfors-holmqvist@med.lu.se
1
Danish Cancer Society Research Center, Copenhagen, Denmark; 2Department of Clinical Medicine, Faculty of Health, Aarhus University, Aarhus, Denmark; 3Institute
for Cancer Outcomes and Survivorship, University of Alabama at Birmingham, Birmingham, Alabama, USA; 4Finnish Cancer Registry, Helsinki, Finland; 5Icelandic
Cancer Registry, Reykjavik, Iceland; 6Faculty of Medicine, University of Iceland, Reykjavik, Iceland; 7Norwegian Cancer Registry, Oslo, Norway; 8Department of Pediatric
Medicine, Oslo University Hospital, Oslo, Norway; 9Institute of Clinical Medicine, University of Oslo, Oslo, Norway; 10Department of Pediatrics, Aarhus University Hospital,
Aarhus, Denmark; 11Pediatric Oncology and Hematology, Skane University Hospital, Lund, Sweden; 12Department of Clinical Sciences, Lund University, Lund, Sweden.
We thank Andrea Bautz for her help with data management and the Adult Life After Childhood Cancer in Scandinavia board (Lars Hjorth, Kirsi Jahnukainen, Nea
Malila, Jørgen H. Olsen, and Catherine Rechnitzer) for their valuable help and guidance.
DOI: 10.1002/cncr.31696, Received: May 24, 2018; Revised: July 14, 2018; Accepted: June 25, 2018, Published online October 11, 2018 in Wiley Online Library
(wileyonlinelibrary.com)
Cancer November 15, 2018
4393
Original Article
CVD subgroups among childhood cancer survivors with
diabetes versus a comparison cohort from the general
population.
MATERIALS AND METHODS
Survivor and Comparison Cohorts
This study is part of the Nordic collaborative study
ALiCCS (https://www.cancer.dk/aliccs/).20 The ALiCCS
study was approved by the national bioethics committee,
the data protection authority, or the national institute for
health and welfare in the respective countries (Denmark,
2010-41-4334;
Finland,
THL/520/5.05.00/2016;
Iceland, VSN 10-041; Norway, 2011/884; and Sweden,
Ö 10-2010, 2011/19).
The basic childhood cancer cohort of ALiCCS includes 43,909 individuals registered with a cancer diagnosis before the age of 20 years in Denmark, Finland,
Iceland, Norway, and Sweden between the start of the
cancer registries in the 1940s and 1950s and December
31, 2008, and it has been described in detail elsewhere.20
For each childhood cancer patient, 5 comparison subjects were randomly selected from the national population registers: they were alive on the date of the cancer
diagnosis of the corresponding childhood cancer patient;
were of the same sex, age, and country (Denmark and
Iceland) or county/municipality (Finland, Norway and
Sweden) of residence; and did not have a cancer diagnosis
before the age of 20 years. For 317 survivors, fewer than 5
comparison subjects were available, and this left 219,131
individuals for the study. Information on vital status and
migration during follow-up was collected from the national population registers for all study subjects.
Before the linkage of study subjects to the national
hospital registers, we excluded those in whom more than
1 primary cancer had been diagnosed in childhood (305
patients), those who had died or emigrated before the
start of the national hospital registers (Sweden, stepwise
inclusion of counties in 1964-1987 and nationwide since
1987; Finland, 1975; Denmark, 1977; Iceland, 1999;
and Norway, 2008; 7251 patients and 5146 comparison
subjects), and those who had died or had been censored
during the first year after the date of the cancer diagnosis
or within an equivalent time lag for the comparison subjects (3193 patients and 1093 comparison subjects). This
resulted in a cohort of 33,160 one-year cancer survivors
and 212,892 comparison subjects.
Hospital Contacts for Diabetes and CVD
The nationwide hospital registers contain information on
virtually all nonpsychiatric hospital admissions in each
4394
of the 5 countries.21,22 Registration is mandatory and
recorded by the treating physician. Each hospital admission initiates a record, which includes the personal identification of the patients, dates of admission and discharge,
a primary discharge diagnosis, and supplementary diagnoses coded according to the International Classification
of Diseases, Seventh Revision (ICD-7), ICD-8, ICD-9, and
ICD-10. All first hospital admissions and hospital-based
outpatient visits with a primary or supplementary discharge diagnosis of diabetes mellitus (ICD-10 codes E10
and E11) and CVD (Table 1) were identified for all study
subjects. The subgroups of ischemic heart disease (ICD10 codes I20-I25) and cerebrovascular disease (ICD-10
codes I60-I69 and G45) were identified within the CVD
group. Diagnostic categories of ICD-7, ICD-8, and ICD9 were adapted to ICD-10 to the extent possible. Study
subject with a hospital contact for diabetes mellitus (60
survivors and 293 comparison subjects) or CVD (371
survivors and 424 comparison subjects) before the date
of cancer diagnosis or the corresponding date for the
equivalent comparison subjects were excluded. Because
the focus of this study was the effect of diabetes on the
risk of CVD, only diabetes events preceding CVD were
included in the analyses. Furthermore, we excluded individuals in whom a congenital malformation of the
circulatory system had been diagnosed (ICD-10 codes
Q20-Q28; 281 survivors and 768 comparison subjects)
and those with a congenital chromosomal abnormality (ICD-10 codes Q90-Q99; 197 survivors and 206
comparison subjects), who might have confounded any
causal association between cancer treatment, diabetes,
and CVD. This resulted in a cohort of 32,251 one-year
survivors and 211,201 comparison subjects. However, because of the probable selection bias regarding survivors
diagnosed with cancer in the earlier decades of the cancer registries and still alive at the start of follow-up in
the hospital registers and because information on their
possible hospital contacts before the hospital registers
became operational was not available, only 1-year survivors diagnosed with cancer in 1968 and onward and
their corresponding comparison subjects were included
in the analyses. Thus, the final cohort included 29,324
one-year survivors and 142,742 comparison subjects. Of
these 1-year survivors, 15,759 (54%) were under continuous follow-up within the hospital registries for 5 years or
more after their cancer diagnosis.
Statistical Analysis
In this longitudinal analysis, follow-up for a first hospital contact for diabetes and CVD started 1 year after the
Cancer November 15, 2018
Heart Disease in Survivors With Diabetes/Winther et al
TABLE 1. ICD-10 Codes for Included
Cardiovascular Diseases
Cardiovascular Disease
ICD-10
Ischemic heart disease
Acute myocardial infarction
Other and chronic ischemic heart
diseases
Pulmonary heart disease
Pulmonary embolism
Pericardial, myocardial, and endocardial
diseases
Acute pericarditis
Other and chronic pericardial diseases
Myocarditis
Endocarditis
Pericarditis
Valvular disease (nonrheumatic)
Mitral valve dysfunction
Aortic valve dysfunction
Other heart disease
Heart failure and congestive heart failure
Congestive heart failure
Cardiomyopathy
Conduction disorders
Atrial fibrillation/flutter
Sick sinus, AV block, BBBs
I20-I25
I21
I20, I22-I25
Supraventricular tachycardia
Other paroxysmal tachycardias
Ventricular tachycardia
Cardiac arrest and/or ventricular
fibrillation
Sick sinus
AV block (1, 2, and 3)
Sick sinus, AV block
Cerebrovascular disease
Infarction
Hemorrhage
Stroke, unspecified
TIA
Subarachnoidal hemorrhage
Infarction and TIA
Occlusion or stenosis of arteria carotis
Occlusion or stenosis of other
precerebral arteries
Arterial disease
Atherosclerosis
Aneurysms (aortic and other)
Arterial embolisms and thrombosis
Venous and lymphatic disease
Phlebitis, thrombophlebitis, and other
venous embolisms and thromboses
Varicose veins, varices, and
hemorrhoids
Pulmonary and venous embolisms
I47.1
I45.6, I47.9
I47.2
I46, I49.0
I26-I28
I26
I30-I33, I38-I39,
I40-I41, I51.4
I30
I31-I32
I40-I41, I51.4
I33, I38-I39
I30-I32
I34-I37
I34
I35
I51.0-3
I42-I43, I50, I51.5, I51.7
I50
I42-I43
I44-I49
I48
I49.5, I44, I45.0-5,
I45.8-9
I49.5
I44.0-3
I49.5, I44.0-3
I60-I69, G45
I63
I61-I62
I64
G45, I66
I60
I63, I66, G45
I65.2
I65.0-1, I65.3-9
I70-I79
I70
I71-I72
I74
I80-I89
I80-I82
I83-I86
I26, I80-I82
Abbreviations: AV, atrioventricular; BBB, bundle branch block; ICD-10,
International Classification of Diseases, Tenth Revision; TIA, transient
ischemic attack.
date of the cancer diagnosis or the corresponding date
for the comparison subjects or at the start of the hospital registers, whichever occurred last. Follow-up ended
on the date of death, emigration, or the end of the study
(Iceland, December 31, 2008; Sweden, December 31, 2009;
Denmark, October 31, 2010; Norway, December 31, 2010;
and Finland, December 31, 2012), whichever occurred first.
Cancer November 15, 2018
The cumulative incidence of CVD with corresponding 95% confidence intervals (CIs) was estimated
with the competing risk of death taken into account with
the Aalen-Johansen estimator.23 The time since cancer
diagnosis and the corresponding date of inclusion for
comparison subjects were used as the underlying time
scale. A stratified Cox proportional hazards model was
used to estimate the effect of a preceding diagnosis of
diabetes and childhood cancer on the cause-specific
hazards of CVD, ischemic heart disease, and cerebrovascular disease. The model was stratified by the age at
diagnosis/inclusion, the year of diagnosis/inclusion, and
sex. The age at cancer diagnosis/inclusion and the year
of diagnosis/inclusion were categorized into 4 groups: 0
to 4, 5 to 9, 10 to 14, and 15 to 19 years and 1968-1977,
1978-1987, 1988-1997, and 1998-2008, respectively.
Schoenfeld residuals were used to test the proportionality assumption of the fitted Cox proportional hazards
models.24 Because of violations of the proportionality assumption when survivors and comparison subjects were
compared, we split the follow-up time of survivors at 15
years after the cancer diagnosis. Thus, we estimated the
effects of being a 1-year childhood cancer survivor in
comparison with a comparison subject on the hazard of
CVD 1 to 15 years after the cancer diagnosis and from
15 years after the cancer diagnosis until the end of follow-up. We thereby ensured that the proportionality assumption was fulfilled. To investigate whether the effect
of diabetes on the hazard of CVD was different among
survivors and comparison subjects, an analysis including
an interaction between diabetes and childhood cancer
was conducted.
To estimate the effects of sex, cancer type, treatment era, and diabetes on the risk of CVD, ischemic
heart disease, and cerebrovascular diseases, we conducted
a multivariate analysis within the survivor cohort. A Cox
proportional hazards model was used to estimate the effects of these variables on the cause-specific hazards of
CVD, ischemic heart disease, and cerebrovascular diseases with the time since cancer diagnosis as the underlying time scale. The models were stratified by the age at
cancer diagnosis (4 age groups as discussed previously).
The following covariates were used: type of cancer (central nervous system [CNS] tumors, leukemia and lymphoma, and solid tumors), treatment era (4 time periods
as discussed previously), and diabetes. In all analyses, the
diagnosis of diabetes was included as a time-dependent
covariate. Accordingly, we estimated the robust standard
errors. Results are presented as hazard ratios (HRs) with
corresponding 95% CIs. R statistical software (version
4395
Original Article
3.4.1) and packages etm and survival were used for the
statistical analyses.
RESULTS
In this cohort of 29,324 one-year survivors of childhood
cancer, 324 and 2108 individuals were diagnosed with
diabetes and CVD, respectively, during follow-up. The
corresponding numbers among the 142,742 comparison
subjects were 1160 and 5275, respectively. Furthermore,
within the survivor cohort, 224 and 443 individuals were
diagnosed with ischemic heart disease and cerebrovascular disease (Table 2).
Figure 1 shows the cumulative incidences of CVD
among survivors and comparison subjects with and without diabetes. The cumulative incidence of CVD among
survivors with diabetes increased more rapidly during the
first 15 years after the cancer diagnosis, whereas the cumulative incidence among the comparison subjects with
diabetes increased slowly the first 20 years after inclusion. However, there was an overlap of CIs.
The hazard of diabetes during follow-up was 1.7
times higher among survivors than comparison subjects
(HR, 1.7; 95% CI, 1.5-1.9), whereas the hazard of CVD
was increased 3.6 times (HR, 3.6; 95% CI, 3.3-3.8) 1 to
15 years after the cancer diagnosis and 1.9 times more
than 15 years after the cancer diagnosis (HR, 1.9; 95%
CI, 1.8-2.0). When we compared the cause-specific hazards of CVD among survivors and comparison subjects
as well as the effect of diabetes, individuals with diabetes
had a 2.4 times higher hazard of CVD (HR, 2.4; 95%
CI, 2.1-2.8) than individuals without diabetes in both
the survivor and comparison cohorts (Table 3). There
was no statistically significant difference in the effect of
diabetes on the hazard of CVD between survivors and
comparison subjects (HR for the interaction of CVD and
diabetes mellitus 1 to 15 years after the cancer diagnosis,
1.2; 95% CI, 0.6-2.4; P = .6; HR for the interaction of
CVD and diabetes mellitus more than 15 years after the
cancer diagnosis, 1.2; 95% CI, 0.8-1.8; P = .3). Thus, for
survivors with diabetes, the HR of CVD was 8.7 (95%
CI, 7.3-10.2) 1 to 15 years and 4.7 (95% CI, 4.0-5.5)
more than 15 years after the cancer diagnosis in comparison with comparison subjects without diabetes.
Within-Cohort Analyses
Table 4 shows the results from the within-cohort analysis
investigating the effect of diabetes as well as the effects
of sex, cancer type, and treatment era on the hazard of
CVD, ischemic heart disease, and cerebrovascular disease.
4396
TABLE 2. Diabetes Mellitus, Cardiovascular Disease,
Ischemic Heart Disease, and Cerebrovascular
Disease in 29,324 One-Year Survivors of Childhood
Cancer and 142,742 Comparison Subjects
Entire cohort
Diabetes mellitus during
follow-up
Cardiovascular disease
during follow-up
Ischemic heart disease
during follow-up
Cerebrovascular disease
during follow-up
Diabetes mellitus before
cardiovascular disease
Diabetes mellitus before
ischemic heart disease
Diabetes mellitus before
cerebrovascular disease
1-y Survivors
No.
Comparison
Subjects, No.
29,324
324
142,742
1160
2108
5275
224
473
443
480
45
91
12
58
15
18
Included cardiovascular diseases are shown in Table 1.
Survivors with diabetes were found to have a 2.9-fold increased risk of CVD in comparison with survivors without
diabetes (95% CI, 2.1-3.9). Furthermore, diabetes increased
the risk of ischemic heart disease and cerebrovascular disease 3.4 (95% CI, 1.9-6.4) and 4.4 times (95% CI, 2.6-7.4),
respectively. Compared with survivors of CNS tumors, survivors of leukemia and lymphoma had an increased hazard
of CVD (HR, 1.2; 95% CI, 1.0-1.3). In contrast, survivors
of solid tumors had a lower risk (HR, 0.8; 95% CI, 0.70.9). Survivors of leukemia and lymphoma had a 3.6 times
increased risk of ischemic heart disease in comparison with
survivors of CNS tumors (95% CI, 2.4-5.4). Survivors of
CNS tumors had an increased risk of cerebrovascular disease in comparison with survivors of leukemia and lymphoma and solid tumors (CNS tumors, referent; leukemia
and lymphoma, HR, 0.4; 95% CI, 0.3-0.5; solid tumors,
HR, 0.2; 95% CI, 0.2-0.3).
There was no significant effect of sex on the risk of
CVD or cerebrovascular disease, whereas female survivors had a significantly lower risk of ischemic heart disease in comparison with male survivors (HR, 0.5; 95%
CI, 0.4-0.7). Compared with survivors diagnosed with
cancer between 1968 and 1977, survivors treated in later
decades had a higher risk of CVD, with an HR of 3.1
(95% CI, 2.6-3.8) for survivors diagnosed with cancer
between 1998 and 2008. We found no significant effect
of treatment era on the risk of ischemic heart disease or
cerebrovascular disease apart from a slight increase in
the risk of ischemic heart disease among those diagnosed
with cancer in 1988-1997 versus 1968-1977 (HR, 1.9;
95% CI, 1.1-3.3).
Cancer November 15, 2018
Heart Disease in Survivors With Diabetes/Winther et al
TABLE 3. HRs of CVD Among 29,324 One-Year
Survivors of Childhood Cancer
Comparison subjects without diabetes
Comparison subjects with diabetes
Survivors without diabetes (HR of CVD
1-15 y after cancer diagnosis)
Survivors without diabetes (HR of CVD
>15 y after cancer diagnosis)
Survivors with diabetes (HR of CVD
1-15 y after cancer diagnosis)
Survivors with diabetes (HR of CVD
>15 y after cancer diagnosis)
Interaction of CVD 1-15 y after cancer
diagnosis and diabetes
Interaction of CVD >15 y after cancer
diagnosis and diabetes
HR
95% CI
P
1.0
2.4
3.6
2.1-2.8
3.3-3.8
<.001
<.001
1.9
1.8-2.0
<.001
8.7
7.3-10.2
<.001
4.7
4.0-5.5
<.001
1.2
0.6-2.4
.6
1.2
0.8-1.8
.3
Abbreviations: CI, confidence interval; CVD, cardiovascular disease;
HR, hazard ratio.
Included CVDs are shown in Table 1.
Figure 1. (A) Cumulative incidence of CVD among 1-year
childhood cancer survivors and comparison subjects with
diabetes by the time since cancer diagnosis/inclusion. The
dashed lines correspond to the 95% confidence intervals.
(B) Cumulative incidence of CVD among 1-year childhood
cancer survivors and comparison subjects without diabetes
by the time since cancer diagnosis/inclusion. CVD indicates
cardiovascular disease; DM, diabetes mellitus.
DISCUSSION
This population-based study of 1-year survivors of childhood cancer shows that survivors with diabetes have a
markedly increased risk of developing CVD in comparison with survivors without diabetes. This increase in risk
is seen throughout the follow-up. However, diabetes does
not increase the risk of CVD to a larger extent among
survivors versus the general population.
In this cohort, 1-year survivors had a 1.7 times increased risk of diabetes in comparison with the general
Cancer November 15, 2018
population, and this is in accordance with earlier reports.9,11,13 The increase in the risk of CVD in the current survivor cohort versus the general population is also
in line with previously published results.5,6 Notably, we
show that the increase in risk for CVD among survivors
was higher the first 15 years after the cancer diagnosis,
although it was still almost 2-fold after 15 or more years.
We have demonstrated that survivors with diabetes have an almost 2.5-fold increased risk of CVD in
comparison with survivors without diabetes. Moreover,
we have shown for the first time that diabetes is not a
more significant risk factor for CVD among survivors
than the general population. In a CCSS study investigating risk factors for major cardiac events, diabetes alone
significantly increased the risk of heart failure and arrhythmia among survivors treated with chest-directed
irradiation.17 However, there was no increase in relative
risk for coronary heart disease or valvular disease in survivors who had received chest-directed radiotherapy or
for heart failure in patients treated with anthracyclines.
Survivors treated with chest-directed irradiation did not
have a statistically significant increase in relative risk
due to the interaction between cardiotoxic exposure and
the acquired risk factor diabetes for the studied CVDs,
except for heart failure. The difference in the distribution of CVD, where those selected in the CCSS (coronary artery disease, heart failure, valvular disease, and
arrhythmia) constituted only a portion of those included
in our study,6 could have contributed at least partly to
the difference in results. Also, unlike the CCSS study,
the analyses of CVD in the current study were not restricted to patients who received cardiotoxic therapy but
included all survivors. Lastly, differences in study design,
4397
Original Article
TABLE 4. Hazard of Risk of Cardiovascular Disease, Ischemic Heart Disease, and Cerebrovascular Disease
Among 1-Year Survivors of Childhood Cancer
Cardiovascular Disease
HRa
Diabetes
Sex
Cancer type
Treatment era
Without diabetes
With diabetes
Male
Female
CNS tumor
Leukemia and
lymphoma
Solid tumor
1968-1977
1978-1987
1988-1997
1998-2008
95% CI
P
2.9
1.0
1.0
1.0
1.2
2.1-3.9
<.001
0.9-1.1
.9
1.0-1.3
.015
0.8
1.0
1.5
1.9
3.1
0.7-0.9
.001
1.3-1.7
1.6-2.3
2.6-3.8
<.001
<.001
<.001
1.0
Ischemic Heart Disease
HRa
Cerebrovascular Disease
HRa
95% CI
P
3.4
1.0
0.5
1.0
3.6
1.9-6.4
<.001
0.4-0.7
<.001
2.4-5.4
<.001
1.1
1.0
1.3
1.9
2.0
0.7-1.7
.7
0.9-1.9
1.1-3.3
0.7-5.7
.2
.032
.2
1.0
1.0
4.4
1.0
0.9
1.0
0.4
0.2
1.0
1.2
1.2
1.2
95% CI
P
2.6-7.4
<.001
0.7-1.0
.101
0.3-0.5
<.001
0.2-0.3
<.001
0.9-1.6
0.8-1.7
0.8-1.8
.2
.3
.5
Abbreviations: CI, confidence interval; CNS, central nervous system; HR, hazard ratio.
Included cardiovascular diseases are shown in Table 1.
a
The analysis was stratified by the age at cancer diagnosis.
where the CCSS included only children diagnosed with
cancer up to 1986, was restricted to 5-year survivors, and
was based on self-report, could have contributed to differences in results between the 2 studies.
Although this study shows that diabetes does not
affect the risk of CVD to a greater extent among survivors than comparison subjects, the risk of CVD among
survivors is, nonetheless, elevated and, therefore, of substantial importance. Because it has been shown that early
diagnosis and improved treatment of diabetes decrease
the risk of later complications of diabetes, including
CVDs, in the general population,16,18,19 it is likely that
early detection of diabetes through screening as well as
intensified efforts for early and optimal treatment could
reduce the risk of complications also among childhood
cancer survivors with diabetes. This could, in turn, lead
to a decline in morbidity and mortality in this vulnerable population. Therefore, screening for and treatment of
diabetes as well as optimal counselling to avoid/postpone
the onset of both diabetes and CVD are essential parts of
the needed long-term follow-up care of childhood cancer
survivors.
A multivariate regression analysis, adjusted for sex,
age at cancer diagnosis, cancer type, and treatment era,
has revealed that survivors with diabetes have an almost 3
times higher risk of CVD than survivors without diabetes. Importantly, diabetes in childhood cancer survivors
increases the risk of ischemic heart disease more than 3
times and the risk of cerebrovascular disease more than 4
times. In the CCSS, diabetes alone did not lead to a significantly higher risk for coronary heart disease among
survivors treated with chest-directed irradiation.17 Again,
4398
the differences in inclusions in the 2 aforementioned
studies may have contributed to the discrepancies in results. In addition, the current study does not include information on the prevalence of cardiovascular risk factors
other than diabetes. In the general population, diabetes
is a well-known risk factor for cerebrovascular disease in
addition to heart disease.25,26 Among childhood cancer
survivors, an increased risk for cerebrovascular disease
has been found, especially among CNS tumor survivors,
predominantly because of irradiation treatment.27-29 In
the CCSS, diabetes was not identified as a risk factor for
stroke in the full childhood cancer survivor cohort, but
diabetes in combination with hypertension had a strong
effect on stroke risk in CNS tumor survivors.30
When one is interpreting the results of this study, it
is important to keep in mind that we used hospital-based
inpatient and outpatient diagnoses of diabetes and CVD.
Although this approach increases the validity of the diagnostic information, patients with diabetes and CVD
exclusively diagnosed by a general practitioner are not
included. However, this limitation also applies to the
comparison cohort. Yet, our results might have been influenced by better medical surveillance of survivors, and
this might have led to potential overestimations of their
risk of diabetes and CVD. Underascertainment of advanced disease seems unlikely. The lack of cancer treatment data hindered investigations of treatment-related
risk factors for diabetes and CVD. Furthermore, separate analyses of type 1 diabetes and type 2 diabetes were
hindered by low numbers. Within ALiCCS, case-cohort
studies of selected outcomes, including diabetes, are
ongoing.
Cancer November 15, 2018
Heart Disease in Survivors With Diabetes/Winther et al
In conclusion, childhood cancer survivors are at increased risk for both diabetes and CVD. This study also
demonstrates that diabetes substantially increases the
risk of CVD, ischemic heart disease, and cerebrovascular
disease in childhood cancer survivors. Although we have
shown that diabetes does not increase the risk of CVD
to a larger extent in survivors than the general population, the substantial increase in the risk of CVD among
survivors with diabetes calls for tailored counselling and
follow-up care, including the early diagnosis and accurate treatment of diabetes, with the aim of decreasing
or preventing the negative impact of diabetes on the development of cardiovascular morbidity and mortality in
long-term survivors of childhood cancer.
FUNDING SUPPORT
This study was funded by the Danish Council for Strategic Research (09066899) and the Swedish Childhood Cancer Foundation (TJ2016-0014).
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
AUTHOR CONTRIBUTIONS
Jeanette F. Winther: Conceptualization, data curation, methodology, formal
analysis, funding acquisition, and writing–review and editing. Smita Bhatia:
Conceptualization, methodology, and writing–review and editing. Luise
Cederkvist: Conceptualization, methodology, formal analysis, and writing–
review and editing. Thorgerdur Gudmundsdottir: Data curation, methodology,
and writing–review and editing. Laura Madanat-Harjuoja: Data curation
and writing–review and editing. Laufey Tryggvadottir: Data curation and
writing–review and editing. Finn Wesenberg: Data curation and writing–review
and editing. Henrik Hasle: Conceptualization, data curation, methodology,
funding acquisition, and writing–review and editing. Anna Sällfors Holmqvist:
Conceptualization, data curation, methodology, formal analysis, funding
acquisition, writing–original draft, and writing–review and editing.
REFERENCES
1. Gatta G, Botta L, Rossi S, et al. Childhood cancer survival in
Europe 1999–2007: results of EUROCARE-5—a population-based
study. Lancet Oncol. 2014;15:35-47.
2. Oeffinger KC, Mertens AC, Sklar CA, et al. Chronic health conditions in adult survivors of childhood cancer. N Engl J Med.
2006;355:1572-1582.
3. Geenen MM, Cardous-Ubbink MC, Kremer LC, et al. Medical assessment of adverse health outcomes in long-term survivors of childhood cancer. JAMA. 2007;297:2705-2715.
4. Kremer LC, van Dalen EC, Offringa M, Ottenkamp J, Voute PA.
Anthracycline-induced clinical heart failure in a cohort of 607 children: long-term follow-up study. J Clin Oncol. 2001;19:191-196.
5. Mulrooney DA, Yeazel MW, Kawashima T, et al. Cardiac outcomes
in a cohort of adult survivors of childhood and adolescent cancer:
retrospective analysis of the Childhood Cancer Survivor Study cohort. BMJ. 2009;339:b4606.
6. Gudmundsdottir T, Winther JF, de Fine Licht S, et al. Cardiovascular
disease in adult life after childhood cancer in Scandinavia (ALiCCS):
a population-based cohort study of 32,308 one-year survivors.
Int J Cancer. 2015;137:1175-1186.
7. Fidler MM, Reulen RC, Henson K, et al. Population-based
long-term cardiac-specific mortality among 34 489 five-year
Cancer November 15, 2018
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
survivors of childhood cancer in Great Britain. Circulation.
2017;135:951-963.
Mertens AC, Liu Q, Neglia JP, et al. Cause-specific late mortality
among 5-year survivors of childhood cancer: the Childhood Cancer
Survivor Study. J Natl Cancer Inst. 2008;100:1368-1379.
Meacham LR, Sklar CA, Li S, et al. Diabetes mellitus in long-term
survivors of childhood cancer. Increased risk associated with radiation therapy: a report for the Childhood Cancer Survivor Study.
Arch Intern Med. 2009;169:1381-1388.
de Vathaire F, El-Fayech C, Ben Ayed FF, et al. Radiation dose to the
pancreas and risk of diabetes mellitus in childhood cancer survivors:
a retrospective cohort study. Lancet Oncol. 2012;13:1002-1010.
van Nimwegen FA, Schaapveld M, Janus CP, et al. Risk of diabetes mellitus in long-term survivors of Hodgkin lymphoma. J Clin
Oncol. 2014;32:3257-3263.
Baker KS, Ness KK, Steinberger J, et al. Diabetes, hypertension, and
cardiovascular events in survivors of hematopoietic cell transplantation: a report from the Bone Marrow Transplantation Survivor
Study. Blood. 2007;109:1765-1772.
Holmqvist AS, Olsen JH, Andersen KK, et al. Adult life after childhood cancer in Scandinavia: diabetes mellitus following treatment
for cancer in childhood. Eur J Cancer. 2014;50:1169-1175.
Soedamah-Muthu SS, Fuller JH, Mulnier HE, Raleigh VS,
Lawrenson RA, Colhoun HM. High risk of cardiovascular
disease in patients with type 1 diabetes in the U.K.: a cohort
study using the general practice research database. Diabetes Care.
2006;29:798-804.
Rosengren A, Vestberg D, Svensson AM, et al. Long-term excess
risk of heart failure in people with type 1 diabetes: a prospective
case-control study. Lancet Diabetes Endocrinol. 2015;3:876-885.
Grundy SM, Benjamin IJ, Burke GL, et al. Diabetes and cardiovascular disease: a statement for healthcare professionals from the
American Heart Association. Circulation. 1999;100:1134-1146.
Armstrong GT, Oeffinger KC, Chen Y, et al. Modifiable risk factors
and major cardiac events among adult survivors of childhood cancer. J Clin Oncol. 2013;31:3673-3680.
Writing Team for the Diabetes Control and Complications Trial/
Epidemiology of Diabetes Interventions and Complications Research
Group. Effect of intensive therapy on the microvascular complications of type 1 diabetes mellitus. JAMA. 2002;287:2563-2569.
Lind M, Svensson AM, Kosiborod M, et al. Glycemic control and excess mortality in type 1 diabetes. N Engl J Med.
2014;371:1972-1982.
Asdahl PH, Winther JF, Bonnesen TG, et al. The Adult Life After
Childhood Cancer in Scandinavia (ALiCCS) study: design and
characteristics. Pediatr Blood Cancer. 2015;62:2204-2210.
Ludvigsson JF, Andersson E, Ekbom A, et al. External review and
validation of the Swedish national inpatient register. BMC Public
Health. 2011;11:450.
Lynge E, Sandegaard JL, Rebolj M. The Danish national patient
register. Scand J Public Health. 2011;39:30-33.
Aalen OO, Johansen S. An empirical transition matrix for non-homogeneous Markov chains based on censored observations. Scand J
Stat. 1978;5:141-150.
Schoenfeld D. Partial residuals for the proportional hazards regression model. Biometrika. 1982;69:239-241.
Tuomilehto J, Rastenyte D, Jousilahti P, Sarti C, Vartiainen
E. Diabetes mellitus as a risk factor for death from stroke.
Prospective study of the middle-aged Finnish population. Stroke.
1996;27:210-215.
Abbott RD, Donahue RP, MacMahon SW, Reed DM, Yano K.
Diabetes and the risk of stroke. The Honolulu Heart Program.JAMA.
1987;257:949-952.
Bowers DC, Liu Y, Leisenring W, et al. Late-occurring stroke among
long-term survivors of childhood leukemia and brain tumors: a
report from the Childhood Cancer Survivor Study. J Clin Oncol.
2006;24:5277-5282.
Campen CJ, Kranick SM, Kasner SE, et al. Cranial irradiation
increases risk of stroke in pediatric brain tumor survivors. Stroke.
2012;43:3035-3040.
4399
Original Article
29. Van Dijk IW, Van der pal HJ, Van Os RM, et al. Risk of symptomatic stroke after radiation therapy for childhood cancer: a
long-term follow-up cohort analysis. Int J Radiat Oncol Biol Phys.
2016;96:597-605.
4400
30. Mueller S, Fullerton HJ, Stratton K, et al. Radiation, atherosclerotic
risk factors, and stroke risk in survivors of pediatric cancer: a report
from the Childhood Cancer Survivor Study. Int J Radiat Oncol Biol
Phys. 2013;86:649-655.
Cancer November 15, 2018