Background
It has been hypothesized that the extent of acute anthracycline-induced cardiotoxicity reflects the risk for late development of heart failure. The aim of this study was to examine if short-term changes in cardiac function can be detected even after low-dose adjuvant epirubicin therapy for breast cancer when using Doppler tissue imaging of longitudinal left ventricular function.
Methods
Eighty consecutive women in good cardiopulmonary health scheduled to undergo adjuvant treatment for breast cancer were included. They were examined using echocardiography and Doppler tissue imaging before and after three treatment series of epirubicin (mean cumulative dose, 273.7 ± 46.6 mg/m 2 ; median time interval, 9 weeks; range, 47–113 days).
Results
Apart from a marginal reduction in E/A ratio, none of the conventional Doppler echocardiographic or Doppler tissue imaging indices of systolic and diastolic function were affected during epirubicin treatment.
Conclusions
In contrast to several previous studies using tissue Doppler and conventional echocardiography, this study did not document relevant short-term effects of low-dose epirubicin treatment on heart function.
Anthracyclines are effective antineoplastic drugs used in the treatment of breast cancer and in many other malignant diseases, including childhood tumors, soft tissue sarcomas, lymphomas, and leukemias. The use of anthracyclines is limited by serious side effects, the major being cardiotoxicity with resulting chronic heart failure (CHF). The mechanism behind the development of anthracycline-induced cardiomyopathy is complex and probably multifactorial. However, the formation of free radicals with oxidative damage seems to play a central role. The risk for the development of CHF is accentuated by several risk factors, the most important being the cumulative anthracycline dose.
The development of CHF is often postponed by several months to years after the end of anthracycline treatment. After high-dose regimens, the risk for heart failure has been reported to approach 20% to 50%, but even during the treatment course, deterioration of heart function can be seen, with reductions in left ventricular ejection fraction (EF) up to 10%. Heart failure is uncommon during low-dose regimens ( Table 1 ), but a risk for the development of CHF seems to persist many years after treatment.
| Study | Population | n | Design | Follow-up time (mos) | Anthracyclines | Cumulative dose (mg/m 2 ) | Modality | Early or late deterioration | CHF |
|---|---|---|---|---|---|---|---|---|---|
| Jurcut et al. (2008) | A | 16 | P | 5 | Pegylated liposomal doxorubicin | 180 | Echo, DTI | SR, E′ | 0 |
| Dodos et al. (2008) | A | 100 | P | 12 | Doxorubicin | 230 | Echo | Early EF, FS, Tei index | 0 |
| Karakurt et al. (2008) | I | 32 | C | 28 | Doxorubicin, daunorubicin | 50–300 | Echo, DTI | E′, E/E′, MPI | 0 |
| Mantovani et al. (2008) | A | 31 | P | 18 | Epirubicin | 200–400 | Echo, DTI | Early SR, E′/A′; late S′ | 0 |
| Nagy et al. (2008) | A | 40 | R | 12 | Doxorubicin, epirubicin | 300 | Echo, DTI | E/A, E′/A′ | 0 |
| Belham et al. (2007) | A | 61 | R | 6 | Doxorubicin | 300 | Echo, DTI | EF, Tei index | 5% |
| Ganame et al. (2007) | I | 13 | P | 3 | Daunorubicin, doxorubicin, idarubicin | 90–225 | Echo, DTI | E, E′, IVRT, SR, ϵ, Tei index | 0 |
| Mercuro et al. (2007) | A | 16 | P | 9–18 | Epirubicin | 200–400 | Echo, DTI | Early SR; late E′, E′/A′ | 0 |
| Corapcioglu et al. (2006) | I | 21 | P | 3 | Doxorubicin, daunorubicin, epirubicin, idarubicin | 240–600 | Echo, MUGA | EF, FS, PFR | 5% |
| Tassan-Mangina et al. (2005) | A | 20 | P | 42 | Doxorubicin | 210 | Echo, DTI | Early E/A, E′; late EF, S′ | 0 |
It has been hypothesized that the extent of acute treatment-induced cardiotoxicity reflects the risk for the late development of heart failure and that impairment of systolic left ventricular function is preceded by subclinical deterioration. This has been supported by studies showing that a decrease in left ventricular EF during treatment predicted the development of CHF. Serial measurements of EF by radionuclide ventriculography (multigated acquisition) or echocardiography are recommended in various treatment guidelines. A recent study, however, indicated that this approach is inadequate to predict late left ventricular impairment. A search has been prompted for a more sensitive monitoring method with prognostic information that can detect the individuals at risk, enabling closer follow-up, early treatment, and possibly the development of prophylactic treatments. Tissue Doppler velocities of the mitral ring have been found in several settings to possess independent and supplementary prognostic information to the standard evaluation of the left ventricular function.
Several groups have studied short-term changes in cardiac tissue Doppler parameters during low-dose anthracycline treatment. Early reductions in the range of 10% to 20% have been described in regional deformation variables, strain rate and strain, and in tissue velocities E′ (early diastolic mitral ring velocity) and S′ (peak systolic mitral ring velocity) ( Table 1 ).
Traditional Doppler velocity measurements of mitral flow E (peak early diastolic flow velocity) and A (peak flow velocity during atrial contraction) waves reflect the diastolic function of the left ventricle but are also affected by filling pressures. In individuals with various cardiovascular diseases, E′ (early diastolic mitral ring velocity) and flow propagation velocity (Vp) are less preload sensitive markers of left ventricular relaxation, and they are inversely correlated with the time constant of isovolumic pressure decay. These parameters have been combined in indices providing information on left ventricular filling pressures.
Our aim was to examine whether subtle short-term changes in cardiac function can be detected even during low-dose adjuvant anthracycline therapy for breast cancer when using Doppler tissue imaging (DTI) of longitudinal left ventricular function. In addition, changes in other Doppler echocardiographic indices were studied.
Methods
Study Population
The protocol was approved by the local ethics committee, and all participants gave their written informed consents.
Calculation of the necessary sample size was based on statistical power >80%; minimal relevant changes of 0.75 cm/sec in tissue velocities, 0.75 mm in maximal systolic displacement, and 5% in EF; spontaneous individual variability in a group of 34 healthy volunteers; and a 5% level of statistical significance. The observed statistical power is displayed in Table 2 .
| Variable | Detection limit | Statistical power |
|---|---|---|
| EF | 5% | 99% |
| E′ | 0.75 cm/sec | 99% |
| A′ | 0.75 cm/sec | 96% |
| S′ | 0.75 cm/sec | 99% |
| Displacement | 0.75 mm | 94% |
| Tei index | 0.05 | 99% |
| Vp | 0.10 m/sec | 63% |
| E/A | 0.10 | 99% |
| E/Vp | 0.1 | 39% |
The subjects were 80 women consecutively recruited among patients who had received curative surgery for primary breast cancer and were scheduled for standard adjuvant chemotherapeutic treatment with a regimen containing epirubicin (cyclophosphamide and epirubicin). Prior treatment with cardiotoxic drugs, radiation therapy, or heart rhythm other than sinus rhythm were exclusion criteria. According to local treatment guidelines, none of the patients had a history of heart disease before epirubicin treatment was planned; hypertension was accepted. All subjects came from the same geographical suburban area. All but one were Caucasian. Demographic and clinical data of the participants are displayed in Table 3 . Participants were examined before and after a total of three series of epirubicin and cyclophosphamide infusions. Standard and tissue Doppler echocardiography was performed as outlined below. Of 85 included individuals, one was excluded because she was previously treated with anthracyclines, two left the study because of recurrence of malignant disease, and two chose not to complete the final examination. Eighty participants completed the study.
| Variable | Value |
|---|---|
| Age (y) | 52 ± 9 |
| Systolic blood pressure (mm Hg) | 133 ± 19 |
| Diastolic blood pressure (mm Hg) | 81 ± 14 |
| Height (cm) | 167 ± 6.4 |
| Weight (kg) | 70.9 ± 14.5 |
| Body mass index (kg/m 2 ) | 25.4 ± 4.8 |
| Cumulated epirubicin dose (mg/m 2 ) | 273.7 ± 46.6 |
| Cumulated cyclophosphamide dose (mg/m 2 ) | 1,767 ± 85 |
| In treatment for hypertension | 23.8% |
| Diabetes | 2.5% |
| Smoking (current or previous) | 41.3% |
| Primary tumor site (right/left) | 40%/60% |
Echocardiography
A single investigator (J.M.A.) performed all echocardiographic studies and subsequent analyses. All studies were stored on an external server for later analyses offline, blinded for all clinical data. All examinations were done in the left lateral decubitus position corresponding to the time of end-expiration. A Vivid 7 ultrasound unit (GE Vingmed Ultrasound AS, Horten, Norway) with a 2.5-MHz probe was used for all examinations. Second-harmonic imaging was applied. EchoPAC PC ’08 (GE Vingmed Ultrasound AS) was used for analysis of the echocardiographic studies.
Echocardiographic studies were performed according to guidelines. Valvular function was screened using standard two-dimensional imaging as well as pulsed-wave and color Doppler imaging. Three cardiac cycles of two-dimensional two-chamber and four-chamber grayscale images were recorded from an apical view. From the apex, pulsed-wave Doppler traces were collected from an area between the tips of the mitral leaflets in the four-chamber view and from the aortic valve in the five-chamber view using continuous-wave Doppler. Also from an apical four-chamber view, color M-mode recordings of diastolic left ventricular filling were obtained as described by Sessoms et al. The excursions of the tricuspid ring corresponding to the free right ventricular wall were assessed using grayscale M-mode images. Color DTI was performed in the apical two-chamber and four-chamber views. Loops of three consecutive beats were stored. The areas depicted included the anterior, posterior, lateral, and septal sections of the mitral ring. The sector was reduced to achieve high temporal resolution. Frame rates > 175 frames/sec were accepted.
Echocardiographic and Doppler Analyses
Chamber Volumes and Left Ventricular EF
Left ventricular volumes and EF were calculated using standard techniques according to a modified Simpson’s method, as were left atrial volumes.
Aortic and Mitral Flow and Tei Index
Ejection time was measured from the Doppler tracings of the aortic valve. The Doppler tracings of the mitral valve were used for the assessment of peak early and late diastolic filling velocities (E and A) and of E-wave deceleration time. The interval from the end of the A wave to the start of the E wave was measured for use in the calculation of the Tei index.
Vp
Vp represents the velocity of the wave front of early diastolic filling of the left ventricle. Vp was calculated as the slope of the color M-mode recording.
Tricuspid Annular Plane Systolic Excursion
Right ventricular function was assessed using tricuspid anterior peak systolic excursion measured on the M-mode tracings of the tricuspid ring.
Tissue Velocities
We used a technique in which longitudinal tissue velocity traces were acquired from DTI loops, fixing the sampling area onto the relevant section of the mitral ring throughout the cardiac cycle ( Figure 1 ). From the DTI recordings, velocity curves were produced from the anterior, posterior, lateral, and septal sections of the mitral ring. A circular sampling area with a diameter of 6 mm was used. Curve smoothing was set to 30 msec. The velocity traces consisted of excursions representing systolic contraction (S′) and early (E′) and late (A′) diastolic dilatation, measured as peak velocities. The maximal systolic displacement, d , was calculated as an integral of the velocity curve during the ejection phase ( Figure 1 ). The tissue Doppler velocity and displacement measurements were averaged over the four sections of the mitral ring.
Statistical Analyses
Values measured from continuous-wave, pulsed-wave, and DTI traces, Vp, and tricuspid anterior peak systolic excursion were averaged over three consecutive beats. Continuous data are presented as mean ± SD for normally distributed data, as medians and ranges for skewed distributions, and as percentages for categorical data. Two-sample paired t tests were used to test changes in the echocardiographic measurements. Linear regression models with analyses for covariance were used for assessing the relation between variables. P values < .05 were considered statistically significant. For multiple comparisons, P values were adjusted according to Bonferroni’s method. Calculations were made using SAS version 9.1 (SAS Institute Inc., Cary, NC) and SPSS version 18.0 (SPSS, Inc., Chicago, IL).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
