Effects of anthracycline chemotherapy and Kuan-Sin-Yin on the spectral indices of arterial pulse waveforms in breast cancer patients

The administered traditional-Chinese-medicine KSY decoction contained seven Chinese herbs: Codonopsis pilosula, Atractylodes macrocephala, Glycyrrhiza uralensis, Poria coco, Astragalus membranaceus, Ligustrum lucidum, and Agastache rugosa.

Before the measurements, the subjects were relaxed and rested for at least 10 min. The environmental temperature was within 23–25°C during the entire measuring period. All subjects gave their informed consent before experiments commenced, were asked to not take any medication for 3 days before experiments, and did not consume food at least 1 h before each experiment. All subjects were non-smokers, and did not take coffee or drinks containing alcohol at least 1 day before experiments [18–20].

For each experiment, the subjects were sitting on a chair, and ECG and BPW signals were measured noninvasively; self-reporting of cancer-related symptoms, Functional Assessment of Cancer Therapy-General (FACT-G; Version.4), Brief Fatigue Inventory-Taiwan (BFI-T), and the EORTC QLQ-C30 (Version.3) questionnaires were applied to the subjects. ECG signals were measured by surface electrodes, and acquired by a preamplifier (lead II, RA-LL; 6600-series, Gould, USA). The BPW signal was acquired by a pressure transducer (KFG-2-120-D1-11, Kyowa) held onto the skin surface above the radial artery 2 cm from the left wrist. The signals were connected to a self-made current-to-voltage converter circuit, and then connected to an analog-to-digital converter card (PCI- 9111DG, Adlink Technology, Taiwan) operating at a sampling rate of 1024 Hz. Before the measurement, the heart rate (HR), brachial systolic BP and diastolic BP were measured by using a sphygmomanometer (MG150f, Rossmax) [18–20].

One thermistor was attached around the wrist to monitor the skin-surface temperature. The resistance of the thermistor was transformed into voltages (by a custom-made circuit) that were also sampled every minute by the analog-to-digital converter card. The acceptable range for the temperature stability during the baseline period was a temperature variation of less than 1.0 °C.

2.2 Analysis

Frequency-domain analysis was applied to derive the 40 harmonic indices from the measured BPW signal (n = 1-10): amplitude proportion (C _n ), coefficient of variation of C _n (CV _n ), phase angle (P _n ), and standard deviation of P _n (P _n _SD) [18–20].

Each individual pulse (between foot points) can be represented by the following finite series. The pulses were excluded if the values between the two foot points were larger than 20% of the pulse amplitude. x ( t ) = A 0 2 + { ∑ n = 1 k / 2 A n cos n ω t s + ∑ n = 1 k / 2 B n cos n ω t s }

The Fourier coefficients (A _n and B _n ) of the pulse can be calculated as A n = 2 k ∑ s = 0 k x s cos n ω t s ( for n = 0 , 1 , . . . , k 2 ) B n = 2 k ∑ s = 0 k x s sin n ω t s ( for n = 0 , 1 , . . . , k 2 ) where ω is the angular frequency and t _s is the sampling time interval.

The amplitude (Amp _n ) and phase angle (P _n ) of each harmonic of the pulse harmonic spectrum can then be calculated as Am p n = A n 2 + B n 2 and P _n  = arctan(B _n /A _n ). The amplitude proportions (C _n values) for each pulse were calculated as Amp _n /Amp_0 ×100%, for n = 1–10. CV_n was then calculated as the coefficient of variations of C _n , and P_n_SD was calculated as the standard deviation of P_n.

Signal processing was performed with MATLAB (MathWorks). The differences were considered significant when p < 0.05; all p-values were two-sided hypotheses. Kruskal– Wallis test (nonparametric ANOVA) was used to compare the indices between groups. Post hoc multiple comparisons tests were made by Dunn’s test.

4.1 Comparisons of pulse indices

For the amplitude proportions (C _n ), Fig. 3 indicates that there were more-prominent changes at M1 in Group NKSY (significant decreases for C₃ and C₅– C₁₀). At M2, C₅– C₁₀ tended to return from the values at M1 to those at M0 (significant for C₅). Changes in C _n at M1 were less prominent in Group KSY than in Group NKSY.

Regarding vessel wall properties, one important characteristic of cancer is angiogenesis [26]. Experimentally induced tumors have been found to contain high levels of vasodilators [27]. Vascular dissemination is a major mechanism by which BC cells migrate into the systemic circulation, leading to distant metastasis and mortality [6]. Associations between adenoma and atherosclerosis have been reported [28]. The mean vascular stiffness has been found to be higher in invasive cancers with poor prognostic features than in cancers with good prognostic features [29]. Acute changes are observed in pulse-wave velocity and distensibility following BC chemotherapy, and these changes partially reverse after the termination of chemotherapy.

Increases in arterial stiffness and endothelial dysfunction have been demonstrated in cancer patients treated with anthracycline chemotherapy [21, 30]. Stiffening of large elastic arteries has been noted in the carotid artery and the aorta [21]. Impairment of the vessel wall has been identified using fluorescence optical imaging [31]. The underlying mechanisms can be attributed to ROS-associated oxidative stress and chronic inflammation [21]. Impaired endothelial function may result in reduced relaxation and increased contraction, and hence increased arterial stiffness [30]. A reduced left ventricular ejection fraction and impairment of vascular function may change the pulse wave propagation and decrease blood supply to peripheral vascular beds, and hence could partly account for the side effects on the quality of life in cancer patients.

KSY has been reported to reduce tumor growth and suppress the proliferation of cancer cells. Analyses of HR variability revealed that KSY induced decreases in the sympathetic neural activity [24, 25]. A study of the microcirculatory blood flow using LDF measurements found significant differences in the approximate entropy and spectral power of blood-flow signals in BC patients receiving chemotherapy between with and without using KSY, which indicates that different microcirculatory-blood-flow responses are induced by KSY administration. Improvement of EORTC QLQ-C30 scores (indicating improvement of cancer-related side effects) was also found in this study, illustrating that restoring the blood-flow supply by administering KSY may help to restore the physiological functioning of BC patients [18]. Skin-surface measurements made following KSY treatment have identified increases in electrical conductivity, which could be related to improvements in the local blood supply [24].

The decreases in C _n at M1 in Group NKSY occurred predominantly in the higher-frequency components. Since the pulse can distend the arterial wall and push blood through the arteriolar openings to supply the tissue, C _n can represent the power of that frequency component within BPW. The main (lower-frequency) frequency components constituting the largest proportion of the BPW power could be more strongly related to pulse transmission in the main artery, with the higher-frequency components being more strongly related to pulse transmission in the peripheral arteries. It is possible that decreases in C₅– C₁₀ are related to the decreased distribution efficiency of the pulse transmission and hence the blood flow perfusion to local tissues. Such changes in C _n can therefore be utilized to monitor impairments of the blood-supply condition in BC patients following chemotherapy. Based on this conjecture, the less-prominent changes in C _n at M1 in Group KSY could be related to the maintenance of the propelling force for the blood flow perfusion following chemotherapy. This illustrates that the possible physiological benefits of using KSY treatment may be partly attributed to the maintenance of the blood supply condition and hence the physiological condition.

The decreases in C₅– C₁₀ at M1 in Group NKSY were slightly restored (i.e., increased; nonsignificant except for C₅) toward the values at M0. It is possible that the harmful effects on the blood flow distribution efficiency could be attenuated at M2, which illustrates the possibility of using the present analysis method to monitor the blood flow distribution condition in BC patients after they receive chemotherapy.

The phase angle can also be obtained by spectral analysis from a time-domain pulse waveform. The P _n value physically corresponds to the time delay to the onset point of the pulse waveform. It can be correlated with the wave propagation velocity of each frequency component within the BPW, and hence it could be another useful index for evaluating changes in vascular properties.

Figure 3 shows that the changes in the main frequency components (lower-frequency components; significant decreases in P₁ and P₃) were more prominent in Group NKSY than in Group KSY. A decreased P _n could be related to the decreased transmission velocity and efficiency of the corresponding frequency component, and hence implies decreased propelling efficiency of the blood supply to local tissues in Group NKSY. There were no such significant differences in the main frequency components in Group KSY, which illustrates that the harmful effects on the blood flow distribution efficiency were alleviated in Group KSY.

4.2 Comparisons of pulse variability indices

The main finding in the present study is the different changing patterns in pulse variability indices between Groups KSY and NKSY. There were no significant changes in CV _n and P _n _SD in Group KSY, whereas there were significant decreases in several variability indices in Group NKSY, which were especially significant for many main (lower) frequency components. Therefore, similarly to C _n and P _n , pulse variability indexes can also be useful in monitoring tumor-related hemodynamic changes.

Increased aortic stiffness has been attributed to structural changes in the vascular matrix and interference in the endothelin regulation of vascular tone [16]. The decreases in CV _n and P _n _SD found in the present study could be related to the impaired regulatory activities in Group NKSY. Tumor vessels differ structurally from normal vessels, having ruffled margins outward and across the vessel lumen that may impair the blood flow and result in hypoxia and hypoperfusion [32]. Immature vessel walls lacking muscle cells and adrenergic innervation have also been noted [33]. The impaired vascular regulatory function may have decreased the microcirculatory regulatory activity and hence decreased the pulse variability indices in Group NKSY. The significant decreases in pulse variability indices occurred mainly in the lower-frequency components. Since these components constitute the largest proportion of the power within the pulse waveform, the present findings for pulse variability indices could be related to the impairment of vascular regulatory mechanisms in the main artery in Group NKSY. In Group KSY, the blood supply and the impairments of the vessel wall could be improved; this may help to maintain the vascular regulatory function, and therefore no significant changes in pulse variability indices were found.

In our previous study [18] we used LDF to measure the microcirculatory blood flow in cancer patients with and without applying KSY. Similar to the present findings for the pulse variability indices, variability indices in the microcirculatory blood flow in Group NKSY were also decreased in that previous study. This illustrates that the upstream (the main artery) and downstream (microcirculation) parts of the vascular system may exhibit similar vascular regulatory responses following the administration of KSY in BC patients.

Figure 4 indicates that HR_CV did not differ among M0– M2 in both groups. HR_CV is correlated with the regulatory activities of the heartbeat, and the pulse variability indices may be more strongly correlated with the vascular condition. The present findings indicate that there were significant differences in the pulse variability indices even when there were no significant differences in the HR variability index. HR_CV was also slightly decreased (although nonsignificantly) at M1 in Group NKSY; this is similar to the changes in the BPW variability indices. These observations further illustrate the similarity in the regulatory mechanisms for the source (the heart) and the downstream (the artery) parts of the cardiovascular system in BC patients receiving chemotherapy.

4.3 Questionnaire results

The scores on the EORTC QLQ-C30, BFI-T, and FACT-G questionnaires in Figs. 5 and 6 reveal that the condition in Group NKSY was worse than that in Group KSY for many aspects of the cancer-related quality of life. It has been previously demonstrated that restoring the local blood supply by administering KSY may help to restore physiological functioning (indicated by improvements in the EORTC QLQ-C30 measures) following chemotherapy [18]. The present findings further illustrate that pulse analysis evaluating the pulse-wave transmission condition has potential in monitoring the different physiological and quality-of-life conditions induced by administering KSY. Compared with LDF measurements (as used in previous studies) that may suffer from interference from motion artifacts, pulse waveform measurements have advantages in practical applications such as being more convenient and user-friendly. The present findings may be meaningful in developing techniques to evaluate the blood supply and physiological conditions following chemotherapy or other treatment strategies in cancer patients.