Background
Cardiotoxicity from anthracyclines or cardiac radiation therapy is detrimental to left ventricular (LV) function. However, the long-term effects on right ventricular (RV) performance are largely unknown. The aim of this study was to investigate the long-term impact of cardiotoxic treatment on RV function among adult cancer survivors.
Methods
Adult lymphoma survivors (LSs) who underwent autologous hematopoietic stem cell transplantation in Norway from 1987 to 2008 were invited to undergo cardiovascular evaluation by echocardiography and cardiopulmonary exercise testing. In total, 274 LSs participated. The mean age was 56 ± 12 years, and the mean follow-up time since lymphoma diagnosis was 13 ± 6 years. Echocardiographic parameters were compared with those of age- and gender-matched control subjects from an existing large Norwegian database. RV systolic dysfunction was indicated by two or more abnormal RV systolic parameters according to current recommendations. LV systolic dysfunction was indicated by LV global longitudinal strain > −17%.
Results
All parameters of RV systolic function were impaired in LSs compared with control subjects ( P < .01 for all). The most pronounced difference was observed for tricuspid annular plane systolic excursion: 22.9 ± 4.1 versus 27.1 ± 4.2 mm. Greater cardiotoxic treatment burden was associated with larger RV functional impairment. Tricuspid annular plane systolic excursion correlated with peak oxygen consumption ( r = 0.23, P = .001). RV systolic performance was associated with LV systolic function ( r = 0.49, P < .001 for tricuspid annular plane systolic excursion vs LV global longitudinal strain), but a greater proportion of patients had LV dysfunction (30.8%) compared with RV dysfunction (6.2%) ( P < .001).
Conclusions
RV systolic function was impaired in LSs. The association between RV and LV function indicates a global, long-term cardiotoxic effect. However, RV dysfunction was less prevalent than LV dysfunction.
Highlights
- •
Little is known about RV function after cardiotoxic exposure in adult cancer survivors.
- •
RV systolic function was impaired in LSs compared with control subjects.
- •
Greater cardiotoxic treatment burden was associated with larger RV functional impairment.
- •
RV dysfunction was less prevalent than LV dysfunction.
Lymphoma survivors (LSs) are at increased risk for cardiovascular complications because of treatment with cardiotoxic drugs, such as anthracyclines, as well as cardiac radiation therapy (RT), including direct and/or scattered irradiation. Anthracyclines dose-dependently induce cardiomyocyte apoptosis, eventually leading to overt heart failure. Cardiac RT causes both micro- and macrovascular damage and is associated with increased risks for valvular disease, myocardial infarction, and heart failure.
Multiple studies have assessed left ventricular (LV) function after cardiotoxic treatment among LSs, showing that LV function is reduced during or shortly after lymphoma treatment as well as in long-term survivors. However, less is known regarding the effects on right ventricular (RV) function. The lack of available data on RV function and the fact that RV function has not been adequately studied in cancer survivors was also acknowledged in a recent expert consensus report on cardiac imaging in adult cancer patients. A subclinical decline in RV function has been observed shortly after completion of anthracycline-containing chemotherapy for breast cancer, but no available data describe RV function in adult LSs. In the general population, RV systolic dysfunction is associated with increased mortality from systolic heart failure, but its prognostic value in cancer survivors is unknown.
In the present study, we aimed to assess the long-term effect of cardiotoxic treatment on RV systolic function in adult LSs treated with autologous hematopoietic stem cell transplantation (auto-HCT).
Methods
Study Design and Participants
Patients were enrolled in a cross-sectional Norwegian multicenter study. The eligibility criteria were auto-HCT treatment for Hodgkin or non-Hodgkin lymphoma in Norway between 1987 and 2008, age ≥ 18 years at auto-HCT, and being alive at the time of the survey. The only exclusion criterion was current treatment for relapsed lymphoma. In total, 274 LSs participated, constituting 69% of all eligible LSs. Patient recruitment has previously been described. Participants and nonparticipants ( n = 125) did not differ with regard to the primary diagnosis (non-Hodgkin or Hodgkin lymphoma), gender, mean cumulative doxorubicin dose, proportion receiving cardiac RT, age at survey, or time from auto-HCT ( P > .05 for all, data not shown). Echocardiographic examinations were performed between March 2012 and March 2014. All participants provided written informed consent. The Regional Committee for Medical and Health Research Ethics approved the study.
Treatment
On the basis of the median cumulative anthracycline dose and a previously reported cutoff for high-dose cardiac RT, we categorized the LSs into four groups: low-dose anthracyclines (<300 mg/m 2 ), higher dose anthracyclines (≥300 mg/m 2 ), anthracyclines and low-dose cardiac RT (≤30 Gy), and anthracyclines and high-dose cardiac RT (>30 Gy). Details concerning treatment regimens have been described elsewhere.
Control Group
We recruited control subjects from a large Norwegian cross-sectional population study (the Nord-Trøndelag Health Study), which includes an echocardiographic database of 1,266 participants without known cardiovascular disease, hypertension, or diabetes mellitus. The control subjects were matched 1:1 with study participants according to age, gender, systolic blood pressure, and body mass index.
Echocardiography
After a minimum of 5 min of rest, patients assumed a left lateral decubitus position for ultrasound examinations. As recommended, examinations were performed using parasternal, apical, and subcostal projections, including dedicated RV views. Ultrasound recordings were obtained using echocardiographic scanners (Vivid 7 or E9; GE Vingmed Ultrasound, Horten, Norway) with standard settings, second-harmonic imaging, and optimal gain and contrast. The frame rates were >90 and >40 frames/sec during tissue Doppler recordings and grayscale imaging, respectively. We obtained at least three consecutive cine loops (more than five for patients in arrhythmia), which were stored for offline analysis using dedicated software (EchoPAC version 112; GE Vingmed Ultrasound).
RV dimensions and functional measurements were obtained as recommended. RV wall thickness was measured at end-diastole from the subcostal view or the parasternal long-axis view. RV basal, mid, and longitudinal diameters and right atrial (RA) area were indexed to body surface area. We measured tricuspid annular plane systolic excursion (TAPSE), RV fractional area change (FAC), RV peak systolic velocity at the lateral tricuspid annular plane (RV S′) by pulsed Doppler tissue imaging, and RV index of myocardial performance (RIMP) by pulsed Doppler tissue imaging at the lateral tricuspid annulus. Peak systolic longitudinal RV strain was measured using two-dimensional speckle-tracking echocardiography. We estimated RV strain (average of six segments, including the interventricular septum) and RV free wall strain (average of three segments, excluding the interventricular septum) in a dedicated four-chamber view.
On the basis of previous studies in healthy individuals, we defined the following cutoff values as indicators for abnormality: TAPSE < 17 mm, RV FAC < 35%, RV S′ < 9.5 cm/sec, RIMP > 0.54, and absolute RV free wall strain < 20%. Lacking a validated global definition of impaired RV systolic function, we considered RV systolic dysfunction to be present when at least two of these parameters were below the cut points. We considered a peak velocity of tricuspid valve regurgitation of >2.8 m/sec to indicate elevated pulmonary arterial pressure. The right atrium was considered enlarged when RA area indexed to body surface area was >10.4 cm 2 /m 2 . LV systolic function was estimated using two-dimensional speckle-tracking echocardiography in the three standard apical image planes to obtain LV global longitudinal strain (GLS) in a 16-segment model. LV GLS > −17.0% was taken to indicate abnormal LV systolic function.
When comparing the numbers of patients with abnormal RV and LV function, in addition to the aforementioned definitions, we also used another set of cutoff values derived from the matched healthy control population. For the latter analyses, abnormality was defined as values below the lower limits of normal (<2 SDs) for TAPSE and LV GLS. As a surrogate for LV end-diastolic pressure, we used the E/e′ ratio, where E is the early transmitral inflow velocity by pulsed Doppler and e′ is the mean of the peak early diastolic velocities at the lateral and septal mitral annulus by Doppler tissue imaging.
A single experienced echocardiographer (K.M.), who was blinded to patient treatment, performed all echocardiographic analyses in the LSs. The control subjects were all examined by an experienced cardiologist (H.D.) using a Vivid 7 scanner. All of the original recordings were reassessed by one investigator (K.M.).
Measurement of Peak Oxygen Uptake
We used an ergometer bicycle for maximal, upright, symptom-limited exercise testing. We used an individualized, stepwise protocol in which the workload was increased every minute to reach the age-, gender-, and weight-adjusted expected maximal load after approximately 10 min. Hemodynamic monitoring and gas exchange analyses were performed simultaneously. We calculated the peak oxygen uptake (V o 2 ) per kilogram per minute as the average oxygen consumption during the last 20 sec of exercise and expressed the peak result as a percentage of the age-, weight-, and gender-adjusted reference values.
Statistical Analysis
Data are presented as mean ± SD, median (range), or number (percentage). We performed all statistical analyses in SPSS version 21.0 (SPSS, Chicago, IL). P values ≤ .05 were considered to indicate statistical significance. Normally distributed, continuous data were compared using one-way analysis of variance and Student’s t tests. The Mann-Whitney U test was used to compare skewed data, whereas χ 2 and Fisher exact tests were used for categorical data. Pearson correlation coefficients were used to measure the relationships between LV systolic function (LV GLS) and different parameters of RV systolic function among the LSs. We assessed inter- and intraobserver reproducibility by repeat analysis (K.M. and K.B.) of echocardiograms in 20 randomly selected patients with interpretable imaging data. We present reproducibility as intraclass correlation coefficients with 95% CIs. We applied linear regression analyses to assess differences in parameters of RV systolic function between the LSs (including all and the treatment subgroups) and the control subjects, with adjustment for E/e′ ratio. TAPSE and RV S′ were also adjusted for RV longitudinal diameter. In the control group, individuals previously diagnosed with cardiovascular disease, hypertension, or diabetes were not invited to participate. For methodologic reasons, we applied the same criteria in the LSs. Consequently, when comparing patients and control subjects, we excluded the LSs ( n = 52) who had been diagnosed with any of these comorbidities before the survey and their associated control subjects to maintain 1:1 matching with control subjects.
We also used linear regression analyses to evaluate the effects of cardiac RT on RV systolic function, using LSs treated with anthracycline-containing chemotherapy alone as the reference group. These analyses were adjusted for patient characteristics (gender, age, and observation time since lymphoma diagnosis), cancer treatment (lifetime doxorubicin exposure), and E/e′ ratio. Again, TAPSE and RV S′ were also adjusted for RV longitudinal dimension. The correlations between peak V o 2 (dependent variable) and RV systolic function parameters were assessed using partial correlations controlling for age and gender. TAPSE and RV S′ were also controlling for RV longitudinal dimension. In LSs without concomitant cardiac RT, any associations between the dose of anthracyclines (continuous variable) and RV systolic parameters (dependent variable) were assessed using a linear regression model adjusted for age, gender, E/e′ ratio, and RV longitudinal dimension (TAPSE and RV S′).
Results
Patient Characteristics
Table 1 presents clinical data and demographics according to treatment group. Ten participants received daunorubicin, and 95% of participants received ≤450 mg/m 2 doxorubicin, with a mean cumulative doxorubicin dose of 316 ± 111 mg/m 2 . Thirty-five percent ( n = 97) had also been treated with cardiac RT. Overt heart failure was diagnosed in 11% of LSs ( n = 29), predominantly caused by the cardiotoxic treatment ( n = 26) and otherwise by coronary heart disease ( n = 3). Smoking habits were similar between the LSs (18%) and control subjects (14%) ( P = .29).
| Variable | All LSs ( n = 274) | AC <300 mg/m 2 ( n = 65) | AC ≥300 mg/m 2 ( n = 112) | AC and cardiac RT ≤30 Gy ( n = 59) | AC and cardiac RT >30 Gy ( n = 38) | P value, AC ( n = 177) vs AC and cardiac RT ( n = 97) |
|---|---|---|---|---|---|---|
| Age at diagnosis (y) | 42 ± 13 | 42 ± 13 | 46 ± 13 | 35 ± 11 | 31 ± 10 | <.001 ∗ |
| Observation since diagnosis (y) | 13 ± 6 | 10 ± 4 | 12 ± 5 | 18 ± 7 | 15 ± 5 | <.001 ∗ |
| Time from diagnosis to auto-HCT (y) | 1.3 (0.2-22.7) | 0.4 (0.2-10.6) | 1.7 (0.3-21.4) | 1.2 (0.2-17.2) | 1.7 (0.6-22.7) | .87 ∗ |
| Non-Hodgkin lymphoma | 78% | 95% | 87% | 71% | 32% | <.001 ∗ |
| Men | 62% | 69% | 62% | 59% | 58% | .36 |
| Body mass index (kg/m 2 ) | 26 ± 4 | 27 ± 3 | 26 ± 4 | 26 ± 4 | 26 ± 5 | .95 |
| Systolic blood pressure (mm Hg) | 131 ± 21 | 137 ± 20 | 132 ± 20 | 129 ± 17 | 121 ± 20 | .001 ∗ |
| Diastolic blood pressure (mm Hg) | 77 ± 10 | 80 ± 10 | 77 ± 10 | 75 ± 9 | 74 ± 11 | .002 ∗ |
| Smoking (current/ever/never) | 18%/40%/42% | 20%/43%/37% | 19%/44%/37% | 19%/44%/37% | 16%/32%/53% | .92 |
| Cancer treatment | ||||||
| Doxorubicin (mg/m 2 ) | 300 (0-775) | 215 (0-275) | 400 (300-520) | 300 (150-775) | 400 (80-729) | .03 ∗ |
| Cyclophosphamide (g/m 2 ) | 4.5 (0-12.3) | 3.6 (0-4.4) | 6.0 (0-10.7) | 5.8 (0-12.3) | 3.0 (0-8.0) | .97 ∗ |
| Cardiac RT (Gy) | 29.75 (19-67) | — | — | 20 (19-30) | 40 (31-67) | — |
| Lines of chemotherapy before auto-HCT (1/2/≥3) | 30%/56%/14% | 61%/37%/2% | 17%/64%/19% | 36%/52%/12% | 5%/69%/26% | .19 ∗ |
| Cisplatinum | 4% | 3% | 5% | 3% | 3% | .67 |
| Bleomycin | 12% | 5% | 9% | 20% | 24% | .002 ∗ |
| Allogeneic HCT after auto-HCT | 7% | 0% | 14% | 2% | 3% | .03 ∗ |
| Comorbidities | ||||||
| Hypertension | 35% | 35% | 37% | 39% | 21% | .49 |
| Diabetes mellitus | 10% | 6% | 10% | 19% | 5% | .19 |
| Hypercholesterolemia | 41% | 39% | 38% | 49% | 42% | .20 |
| Thyroid disease | 14% | 9% | 9% | 15% | 34% | .002 ∗ |
| Laboratory parameters | ||||||
| Hemoglobin (g/dL) | 14.0 ± 1.3 | 14.0 ± 1.2 | 13.9 ± 1.4 | 14.0 ± 1.3 | 14.1 ± 1.1 | .47 |
| Creatinine (μmol/L) | 82 ± 22 | 80 ± 16 | 85 ± 28 | 81 ± 18 | 79 ± 18 | .31 |
| N-terminal pro–brain natriuretic peptide (pmol/L) | 17 (1-538) | 17 (2-303) | 16 (1-538) | 14 (2-321) | 35 (2-379) | .07 |
| Glycated hemoglobin (%) | 5.8 ± 0.7 | 5.7 ± 0.5 | 5.8 ± 0.7 | 6.0 ± 1.0 | 5.6 ± 0.5 | .49 ∗ |
| Total cholesterol (mmol/L) | 5.4 ± 1.2 | 5.4 ± 1.1 | 5.5 ± 1.2 | 5.4 ± 1.1 | 5.2 ± 1.2 | .29 |
| Low-density lipoprotein cholesterol (mmol/L) | 3.4 ± 1.1 | 3.4 ± 1.0 | 3.5 ± 1.0 | 3.4 ± 1.1 | 3.3 ± 1.2 | .48 |
| Current medication | ||||||
| Cardioactive medication | 17% | 11% | 16% | 29% | 11% | .11 ∗ |
| Statin | 14% | 12% | 8% | 24% | 18% | .006 ∗ |
| β-blocker | 8% | 6% | 10% | 7% | 5% | .50 |
| ACE inhibitor/ARB | 11% | 9% | 6% | 25% | 8% | .005 ∗ |
| Calcium channel blockers | 4% | 2% | 4% | 10% | 0% | .18 ∗ |
∗ P < .05 from one-way analysis of variance or Kruskal-Wallis test for comparisons between treatment subgroups.
RV Systolic Function in LSs and Control Subjects
Among the LSs, measurement feasibility was 99% for TAPSE, 91% for RV FAC, 79% for RV free wall strain, 96% for RV S′, and 77% for RIMP. On average, all parameters of RV systolic function were impaired in LSs compared with control subjects. The most pronounced difference was observed for TAPSE (22.9 ± 4.1 vs 27.1 ± 4.2 mm, P < .001) ( Table 2 ). Compared with matched healthy control subjects, all predefined treatment groups had lower values for all RV function parameters, except RV S′ ( P < .01 for all). RV S′ was significantly lower only in the group treated with high-dose cardiac RT. Overall, more severe cardiotoxic treatment burden was associated with larger impairment in RV functional parameters ( Table 3 ).
| Variable | LSs ‡ ( n = 222) | Control subjects ( n = 222) | AC <300 mg/m 2 ( n = 65) | AC ≥300 mg/m 2 ( n = 112) | RT ≤30 Gy ( n = 59) | RT >30 Gy ( n = 38) |
|---|---|---|---|---|---|---|
| RV global function | ||||||
| RIMP | 0.46 ± 0.08 † | 0.38 ± 0.06 | 0.47 ± 0.09 † | 0.47 ± 0.09 † | 0.48 ± 0.12 † | 0.48 ± 0.17 † |
| RV systolic function | ||||||
| TAPSE (mm) | 22.9 ± 4.1 † | 27.0 ± 4.2 | 23.8 ± 4.5 † | 23.0 ± 4.0 † | 22.3 ± 4.1 † | 20.8 ± 4.4 † |
| RV S′ (cm/sec) | 12.4 ± 2.5 ∗ | 12.9 ± 2.3 | 13.0 ± 2.8 | 12.3 ± 2.5 | 12.3 ± 2.6 | 10.8 ± 2.2 † |
| RV FAC (%) | 44 ± 5 † | 48 ± 5 | 45 ± 5 † | 44 ± 5 † | 43 ± 5 † | 41 ± 5 † |
| RV strain (%) | −22.8 ± 2.9 † | −25.3 ± 2.3 | −23.3 ± 3.1 † | −22.8 ± 3.0 † | −21.7 ± 3.1 † | −21.1 ± 3.6 † |
| RV free-wall strain (%) | −27.1 ± 4.0 † | −30.0 ± 2.6 | −27.4 ± 4.3 † | −27.3 ± 3.9 † | −26.4 ± 3.9 † | −25.0 ± 3.8 † |
| RV morphology | ||||||
| RV midwall thickness (mm) | 2.9 ± 0.7 † | 3.3 ± 0.7 | 3.0 ± 0.7 † | 2.9 ± 0.6 † | 2.9 ± 0.7 † | 2.7 ± 0.7 ∗ |
| RV end-diastolic area (cm 2 /m 2 ) | 11.1 ± 2.3 ∗ | 11.8 ± 1.9 | 11.4 ± 2.5 | 11.0 ± 2.3 † | 11.2 ± 2.5 | 11.1 ± 2.3 |
| RV basal diameter (cm/m 2 ) | 1.98 ± 0.25 | 2.02 ± 0.24 | 1.98 ± 0.25 | 2.03 ± 0.24 ∗ | 1.89 ± 0.27 | 1.95 ± 0.27 |
| RV mid diameter (cm/m 2 ) | 1.48 ± 0.19 | 1.52 ± 0.17 | 1.48 ± 0.19 | 1.51 ± 0.18 † | 1.41 ± 0.21 | 1.49 ± 0.22 |
| RV long diameter (cm/m 2 ) | 3.9 ± 0.4 | 4.0 ± 0.3 | 3.9 ± 0.4 | 3.9 ± 0.4 | 3.8 ± 0.3 | 3.9 ± 0.4 |
| RA area (cm 2 /m 2 ) | 7.8 ± 1.7 | 7.8 ± 1.3 | 8.0 ± 1.8 | 8.0 ± 2.0 | 7.8 ± 2.0 | 7.6 ± 1.4 |
| LV parameters | ||||||
| LV GLS (%) | −18.6 ± 2.3 † | −20.7 ± 1.9 | −19.0 ± 2.4 † | −18.5 ± 2.8 † | −17.4 ± 2.6 † | −16.6 ± 2.5 † |
| E/e′ ratio | 7.9 ± 3.4 † | 6.9 ± 2.7 | 7.7 ± 3.5 | 8.4 ± 3.6 ∗ | 8.0 ± 3.2 † | 8.3 ± 2.7 ∗ |
∗ P < .05 and † P < .01 compared with control subjects by independent Student’s t test.
‡ LSs ( n = 52) with known cardiovascular disease, hypertension, or type 2 diabetes mellitus before the study were excluded from analyses concerning all comparisons (for all LSs vs control subjects and subgroups) with control subjects.
| Variable | AC <300 mg/m 2 ( n = 56) | AC >300 mg/m 2 ( n = 91) | RT ≤30 Gy ( n = 42) | RT >30 Gy ( n = 33) | ||||
|---|---|---|---|---|---|---|---|---|
| β | 95% CI | β | 95% CI | β | 95% CI | β | 95% CI | |
| TAPSE (mm) | ||||||||
| Unadjusted | −3.1 † | −4.7 to −1.4 | −3.5 † | −4.6 to −2.3 | −4.1 † | −5.9 to −2.2 | −7.1 † | −9.1 to −5.1 |
| Adjusted | −3.7 † | −5.2 to −2.2 | −3.1 † | −4.2 to −1.9 | −4.0 † | −6.0 to −2.0 | −6.8 † | −9.0 to −4.6 |
| RV S′ (cm/sec) | ||||||||
| Unadjusted | −0.4 | −1.4 to 0.6 | −0.3 | −1.0 to 0.4 | −0.1 | −1.1 to 1.0 | −2.1 † | −3.1 to −1.0 |
| Adjusted | −0.5 | −1.5 to 0.5 | −0.3 | −1.0 to 0.4 | −0.4 | −1.5 to 0.8 | −1.6 † | −2.7 to −0.5 |
| RV FAC (%) | ||||||||
| Unadjusted | −4 † | −6 to −2 | −4 † | −5 to −2 | −4 † | −6 to −2 | −7 † | −9 to −4 |
| Adjusted | −4 † | −6 to −2 | −4 † | −6 to −3 | −4 † | −6 to −2 | −6 † | −9 to −3 |
| RV strain (%) | ||||||||
| Unadjusted | −3.5 † | −5.4 to −1.6 | −1.7 † | −2.8 to −0.6 | −2.3 † | −4.0 to −0.7 | −3.8 † | −5.6 to −2.0 |
| Adjusted | −3.5 † | −5.4 to −1.6 | −1.6 † | −2.7 to −0.5 | −2.2 ∗ | −4.0 to −0.4 | −4.1 † | −6.0 to −2.1 |
| RV free wall strain (%) | ||||||||
| Unadjusted | −4.3 † | −6.8 to −1.8 | −2.0 † | −3.4 to −0.5 | −2.3 ∗ | −4.5 to −0.2 | −4.4 † | −6.5 to −2.2 |
| Adjusted | −4.4 † | −6.9 to −1.8 | −1.9 † | −3.4 to −0.4 | −2.4 ∗ | −4.7 to −0.1 | −4.9 † | −7.2 to −2.6 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
