Methods

The study patients participated in a cross-sectional study including outpatient clinical examination of long-term survivors of CL. The present study assessed cardiac function in the patients, with comparison with a matched control group. Patients were identified from the Cancer Registry of Norway. Patients were eligible if they were diagnosed with either non-Hodgkin lymphoma or Hodgkin lymphoma when aged <18 years, between 1970 and 2000, were aged >18 years at the time of the study, and had survived at least 5 years after diagnosis. Whereas the study comprised survivors of Hodgkin lymphoma from all parts of Norway, survivors of non-Hodgkin lymphoma were included from the south-eastern health region of Norway, which covers approximately 50% of the Norwegian population, because of logistic limitations.

Data on treatment methods and doses were obtained from the patients’ medical records. Anthracycline doses were converted to doxorubicin isotoxic doses using a conventional conversion factor of 0.67 × epirubicin dose and 0.833 × daunorubicin dose, respectively. Anthracycline doses are expressed as cumulative dose adjusted for body surface area (m ² ). Total radiation dose to the mediastinum was registered for each patient. Patients with radiation fields not involving the mediastinum were classified as not receiving radiation to the heart. Patients were categorized into the following 4 treatment groups: (1) “anthracyclines only” (i.e., anthracycline treatment without mediastinal radiotherapy), (2) “mediastinal radiotherapy only” (i.e., radiation fields involving the mediastinum without anthracycline treatment), (3) “anthracyclines and mediastinal radiotherapy in combination,” and (4) “no cardiotoxic treatment” (i.e., neither anthracyclines nor mediastinal irradiation). Analyses were also performed on differences between survivors of CL based on whether they had received mediastinal radiotherapy and anthracyclines. In addition, the effect of different radiation doses on valvular dysfunction was analyzed.

All participants underwent an extensive medical assessment including physical examination, blood sampling, electrocardiography, and a comprehensive echocardiographic examination. Written informed consent was given by all the participants. The study complied with the Declaration of Helsinki and was approved by the Regional Committee for Medical and Health Research Ethics.

A control group for echocardiographic parameters was obtained by matching patients 1:1 for gender, age, body weight, and systolic blood pressure to healthy control subjects selected from a clinical echocardiographic database in the third wave of the Nord-Trøndelag Health Study. This is a cross-sectional study of 50,839 inhabitants (54% of the total) from the mid region of Norway, where a complete echocardiographic examination was performed in 1,296 subjects without known cardiovascular disease, diabetes, or hypertension. Matching 1:1 made it possible to compare different treatment groups with their respective controls, as there were some differences in patient characteristics, for example, age at survey, among the groups of survivors of CL based on treatment.

All echocardiograms were obtained from 2007 to 2009 by experienced sonographers according to a standardized protocol using digital high-end echocardiographic scanners (Vivid 7 or E9; GE Healthcare Vingmed Ultrasound, Horten, Norway). The patients were examined at the Department of Cardiology, Oslo University Hospital, and the results of their examination were reviewed by an experienced cardiologist blinded to the patients’ clinical status (SA). The control subjects were examined and reviewed at the Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, by another experienced cardiologist (HD).

LV wall thickness, internal dimension, and fractional shortening (FS) were obtained from M-mode parasternal recordings following convention. LV ejection fraction (EF) and volumes were calculated using Simpson’s biplane method. Cardiac output was calculated from LV outflow tract dimension measured in parasternal long-axis view and the integral of the LV outflow tract pulse-wave Doppler signal. To correct for differences in body size, dimensions and cardiac output were indexed to body surface area. Left atrial dimensions were obtained in the patients only, where diameter was measured by M-mode from parasternal long-axis view and biplane area was measured from apical 4-chamber and 2-chamber views. Systolic LV myocardial longitudinal strain measurement (i.e., regional myocardial shortening) was also obtained in the patients, using semiautomatic software from GE (EchoPAC, version 7) from the standard 3 apical imaging planes, using gray-scale cineloop imaging with a frame rate of at least 40 frames/s. Global peak systolic strain was calculated.

Mitral valve inflow was recorded at the tips of the valve. Pulmonary vein inflow was obtained in the upper right pulmonary vein with the sample volume at least 0.5 cm into the vein. From mitral inflow, peak early (E) and late diastolic velocities were recorded, as well as the deceleration time of the E wave. From pulmonary vein inflow, peak systolic, diastolic, and atrial reversed velocities were recorded. The isovolumic relaxation time was derived from simultaneous Doppler recording of the aortic ejection and mitral inflow. Pulse-wave tissue Doppler velocities were recorded from the mitral annular septal and lateral insertions and the peak systolic and early diastolic (e′) tissue velocities obtained over at least 3 cardiac cycles. Ratios of diastolic mitral inflow velocities, pulmonary vein velocities, and E/e′ (using the average of septal and lateral e′) were calculated.

Normal values for LV size and function refer to published recommendations from the American Society of Echocardiography and European Association of Echocardiography. However, we chose to define EF <50% as abnormal, in accordance with clinical practice. We defined systolic dysfunction as EF <50% or FS <27% in female patients and <25% in male patients. We defined diastolic dysfunction as septal e′ <0.08 m/s or lateral e′ <0.10 m/s.

Valvular disease was graded following convention : regurgitations and stenoses were graded as mild, moderate, or severe. Minimal regurgitations on left-sided cardiac valves and mild regurgitations on right-sided cardiac valves were considered normal. For patients with previous cardiac valve surgery, the valvular dysfunction before surgery was used for the analyses of valvular disease. In patients with tricuspid regurgitation, peak pressure gradient was measured.

Data with normal distribution are expressed as mean ± SDs, whereas data with skewed distribution are presented as median (twenty-fifth to seventy-fifth percentiles). Student t test, Mann-Whitney test, or 1-way analysis of variance test was used to compare differences between groups, as appropriate. Chi-square test was used to compare categorical variables. A 2-sided p value of <0.05 was considered significant. The least significant difference test was used for post hoc analysis of differences between groups detected by 1-way analysis of variance. Logistic regression analyses were used to evaluate the association between the occurrence of cardiac abnormalities (LV systolic and diastolic dysfunction, left heart valvular dysfunction) and gender, diagnosis, age at diagnosis, age at examination, mediastinal radiotherapy, and anthracycline treatment. Independent variables resulting in p <0.2 were included in multivariate analysis. Statistical analyses were performed with PASW Statistics 18 software from SPSS Inc. (Chicago, Illinois).

Results

In all, 220 survivors of CL were eligible. Of these, 95 patients did not participate: 15 answered questionnaires but did not attend examinations, and 80 did not respond to the invitation. Thus, a total of 125 patients (57%) were included and completed the echocardiographic examination ( Table 1 ). The Regional Committee for Medical and Health Research Ethics did not allow further investigation of nonresponders.

Patient characteristics and treatment methods are listed in Table 1 . Patients and control subjects were well matched, with no differences in median age (33 vs 33 years, p = 0.928), mean systolic blood pressure (128 vs 128 mm Hg, p = 0.754), body weight (74 vs 75 kg, p = 0.810) or body surface area (1.87 vs 1.89 m ² , p = 0.581).

Occurrence of LV systolic dysfunction did not differ between patients and control subjects, but diastolic dysfunction occurred 7 times more frequently in patients ( Table 2 ). No associations emerged between patient characteristics or treatment, respectively, and the occurrence of LV systolic dysfunction in regression analyses. Univariate analysis showed a borderline significant association between mediastinal radiotherapy and diastolic dysfunction (odds ratio 2.2, 95% confidence interval 1.0 to 5.0, p = 0.05). In multivariate analysis, however, only age at examination was independently associated with diastolic dysfunction (odds ratio 1.1, 95% confidence interval 1.0 to 1.2, p = 0.001). In multivariate analysis, mediastinal radiotherapy (odds ratio 27.8, 95% confidence interval 6.6 to 118, p <0.001) was independently associated with the prevalence of left-sided valvular dysfunction, in addition to age at examination (odds ratio 1.2, 95% confidence interval 1.1 to 1.3, p = 0.001).

The patients had lower FS than control subjects, except for the no cardiotoxic treatment group ( Table 3 ). Cardiac index was also lower, despite a slightly higher heart rate in patients compared with control subjects (69 ± 12 vs 66 ± 10 beats/min, p = 0.010). EF was not different among treatment groups or between patients and control subjects. Patients treated with mediastinal radiation only had a trend toward lower global strain than the other 3 treatment groups. Comparable strain values were not available in the control group. There was a highly significant, yet small, difference in septal peak systolic mitral annular velocities between patients and controls, whereas lateral systolic velocities were not significantly different.

The patients had reduced diastolic function, as shown by lower e′ and higher E/e′ ratio, than control subjects ( Table 4 ). However, the 9 patients with no cardiotoxic treatment did not differ in lateral e′ and E/e′ compared with control subjects ( Table 5 ). Patients treated with mediastinal radiotherapy only had lower e′, higher E/e′, and higher peak tricuspid valve regurgitation pressure gradient than the other patients ( Table 4 ). After exclusion of patients with left-sided valvular dysfunction from the analyses, there were still highly significant differences in e′ and E/e′ ratio between patients and controls (all comparisons p ≤0.001). The left atrial area was equal in the different patient groups but was not measured in controls ( Table 4 ). Patients treated with mediastinal radiotherapy had lower indexed LV internal dimension than control subjects ( Table 5 ). There were no systematic differences in LV wall thickness.

Stay updated, free articles. Join our Telegram channel

Join