The effect of cardiac resynchronization therapy on arterial stiffness and central hemodynamic parameters

Abstract Introduction Cardiac resynchronization therapy (CRT) is a device‐based method of treatment which decreases morbidity and mortality in heart failure with reduced ejection fraction (HFrEF). This study was aimed to investigate the effects of CRT on hemodynamic and arterial stiffness parameters evaluated by noninvasive method, and determine whether there is a correlation between the changes after CRT in these parameters and the clinical response to CRT or not. Methods The study included 46 patients with HFrEF who were planned to undergo CRT implantation. Before the CRT implantation, clinical and demographic data were recorded from all patients. Hemodynamic and arterial stiffness parameters were measured oscillometrically by an arteriograph before CRT implantation. The patients were re‐evaluated minimum three months after CRT; the above‐mentioned parameters were measured again and compared to the pre‐CRT period. Results Compared to the period before CRT, mean systolic blood pressure (SBP) (116.8 ± 19.1 mm Hg vs 127.7 ± 20.9 mm Hg, P = .005), central SBP (cSBP) (106.2 ± 17.3 mm Hg vs 116.8 ± 18.7 mm Hg, P = .015), cardiac output (CO) (4.6 ± 0.8 lt/min vs 5.1 ± 0.8 lt/min, P = .002), stroke volume (65.6 ± 16.3 mL vs 72.0 ± 14.9 mL), and pulse wave velocity (PWV) (10 ± 1.6 m/sec vs 10.4 ± 1.8 m/sec, P = .004) increased significantly in post‐CRT period. In addition, the same parameters were significantly increased post‐CRT period in patients with clinical response. However, there was not any similar increase in nonresponder patients. Conclusion This study demonstrated that SBP, CO, and PWV increased significantly after CRT. The modest increases in these parameters were observed to be associated with positive clinical outcomes.


| INTRODUC TI ON
Heart failure (HF) is a frequently seen clinical syndrome.
Approximately 1%-2% of the adult population in developed countries have HF, with the prevalence rising to ≥10% among persons 70 years of age or older. 1 Heart failure can be classified into three types based on the condition of left ventricular (LV) systolic function: HF with preserved ejection fraction (EF) (EF ≥50%), mid-range HF (EF: 40%-49%) and HF with reduced ejection fraction (HFrEF) (EF <40%). 2 In HFrEF, CRT which can be applied in selected patients in addition to medical therapy has evidence-based positive effects both on symptoms and prognosis. 3 Arterial stiffness or reduced aortic distensibility has an important role in patients with HF, and is valuable in prognosis. [4][5][6][7][8] Pulse wave velocity (PWV) is the gold standard method of arterial stiffness measurement. 9 PWV is usually measured using carotid-femoral and brachial-ankle methods. 9 Many clinical studies have shown that PWV measurement can predict the clinical prognosis in cardiovascular diseases. [10][11][12][13] However, recently, more feasible PWV measurement became possible with the use of ankle blood pressure monitorization devices. 14,15 When contraction synchronization was provided by CRT, improvement in hemodynamic parameters such as cardiac output and a modest increase in systolic blood pressure (SBP) and pulse pressure (PP) is expected. [16][17][18] However, the impact of CRT on PWV has yet to be investigated. Our study aimed to evaluate the effect of CRT on hemodynamic parameters and PWV by measuring these parameters using a noninvasive, oscillometric method. Additionally, we aimed to detect whether there was an association between these parameters and clinical response to CRT.

| The measurement of arterial stiffness and hemodynamic parameters
Mobil-O-Graph 24 hours ABPM NG® arteriograph was used to measure arterial stiffness and cardiovascular hemodynamic parameters. All measurements were performed by a single experienced operator (medical doctor) who was blinded to other data.
Each subject rested in a seated position for 10 minutes in a quiet room at 20°C-23°C before the baseline hemodynamic measurements were recorded. Brachial blood pressure (BP) and heart rate (HR) were measured in the right arm before using arteriograph.
Three sequential measurement sessions, separated by a 5-min interval were obtained. A total of six measurements, one of which was a control, were taken automatically by the device for a session. The mean values of each session were recorded. If the mean PWV difference between sessions was less than 0.5 m/sec, the mean was used for the analysis. Otherwise, a third measurement session was to be conducted and the median of the three measurements was calculated. Thus, biases caused by single investigator measurements were eliminated. Also the measurements at follow-up were performed at the same time of the day to minimize the variation as proposed by consensus. 19 The measurement was performed after proper cuff size was selected (two sizes available: 24-34 and 32-42 cm). Briefly, underlying working principle of the oscillometric method is; to record oscillometric pressure curves based on plethysmography and register pulsatile pressure changes in an artery on the upper arm.
After the conventional oscillometric BP measurement, the cuff instantly re-inflates and the pulse waves are recorded at diastolic blood pressure (DBP) level for 10 seconds. By this method, the central aortic pressure is calculated from the brachial BP using a transfer function (TF) and the central aortic waveform is decomposed into forward and reflected waves using an uncalibrated triangular aortic flow waveform. The idea behind a TF is the modification of a certain frequency range within the acquired pulse signal to derive the aortic pressure wave (ARCSolver algorithm). The waves formed by central pressure changes amplified and transferred to device-specific tonometry and uploaded to HMS Client Server 5.1® software which was developed specifically for the device. Mobil-O-Graph 24 hours PWA Monitor uses these mechanisms for obtaining measurements which are validated. 14,15,20,21 ( Figure 1).

| Echocardiography
Transthoracic echocardiography (TTE) was performed by an experienced echocardiography specialist who was blinded to other data.

| CRT implantation
Following left pectoral region incision, subclavian venous puncture was performed, and right ventricle and right atrium leads were placed. After this, coronary sinus was found using a left amplatz catheter, and images were recorded using a contrast-enhancing agent for the selection of the suitable branch. Left ventricular (LV) lead was placed on the lateral or posterolateral branch of coronary sinus if possible. All electrodes were connected to the generator, and the pouch was closed following stimulus and threshold values were controlled. After implantation AV delay of the patients was set to be 120 ms, and VV delay was 0 ms for optimal resynchronization.

| Follow-up the patients
The patients were invited for follow-up minimum three months after CRT implantation. The mean follow-up period of the patients was 20 weeks. They were questioned for their compliance to medical therapy and whether they were hospitalized for HF. The information whether they had a dose change in their medications during followup period was noted. Their body weights and heights were measured again, and the BMI was re-calculated. Their functional capacities F I G U R E 1 Evaluation of aortic pulse wave velocity using a transfer function from brachial pressure wave analysis were evaluated according to the NYHA classification, 6MWT was performed. Electrocardiography (ECG) was taken. Arterial stiffness and cardiovascular hemodynamic parameters were measured again.
Pacemaker was controlled, and the ventricular pace rate being 95% and/or above was recorded.

| Definitions
Experts consider aortic pulse wave velocity as the gold standard in the assessment of arterial stiffness. 9 Therefore, we considered PWV as a primary arterial stiffness marker in our study.
Ischemic cardiomyopathy and nonischemic cardiomyopathy definitions were made based on the presence or absence of myocardial infarction events or 75% or more stenosis in the left coronary artery.
Patients with 50-meter and/or 20% increase in 6MWT after CRT, and who were not hospitalized due to decompensation of heart failure during the follow-up were considered as "clinical responders." Those who do not meet these criteria considered as "nonresponders."

| Statistical analysis
Study data were computerized using "SPSS (Statistical Package for Social Sciences) for Windows 21.0 (SPSS Inc)" and assessed.
Descriptive statistics were expressed as mean ± standard deviation, frequency distribution, and percentage. The distribution normality of the variables was examined using visual (histogram and probability graphs) and analytic methods (Shapiro-Wilk Test). For the normally distributed variables Student's T Test was used for the statistical significances between two independent groups, and Paired Sample T Test for the statistical significances between two dependent groups. For not normally distributed data, as statistical methods Mann-Whitney U Test were used between two independent groups and Wilcoxon signed-rank test were used between two dependent groups. Statistical significance level was considered as P value < .05. 32.6% of the patients were smokers, and all patients were receiving optimal medical therapy (including beta-blocker, ACEI/ARB, spironolactone). 54.3% of the patients under assessment were in NYHA class II, 32.6% were in NYHA class III, and 13.0% were in NYHA class III/IV. The mean LVEF of the patients was 28.1% (Table 1). Table 1 showed other demographic and baseline clinical characteristics of patients. 32.6% of the patients were hospitalized due to HF decompensation and two patients (4.3%) died during the follow-up period. One patient died due to septic shock after pneumonia and the other patient died due to retroperitoneal bleeding caused by warfarin overdose. Table 2  hemodynamic parameters. Heart rate was similar between two groups. Post-CRT 6MWT was significantly increased compared to pre-CRT period, and QRS duration was significantly decreased (P < .05). When cardiovascular hemodynamic parameters were assessed, post-CRT SBP, mean arterial pressure (MAP), PP, cSBP, stroke volume, and CO were significantly increased compared to pre-CRT values (P < .05). Also, PWV was significantly increased compared to pre-CRT values (P < .05), no statistically significant change was observed in AIx (P > .05) (  Table 3 showed demographic and baseline clinical characteristics of responder and nonresponder patients. Table 4 shows the distribution of 6MWT, QRS duration, cardiovascular hemodynamic and arterial stiffness parameters of the responders and nonresponders. When intra-group assessment was performed for the clinical responders; post-CRT SBP, DBP, MAP, cSBP, cDBP CO, and PWV were significantly increased compared to pre-CRT values (P < .05). No statistically significant difference was detected in the same parameters for nonresponders (P > .05) ( Table 4; Figures 4 and 5).

| D ISCUSS I ON
This study showed that CRT increased SBP, PP, and CO using a noninvasive oscillometric method. In addition, while we observed a significant increase in these parameters in patients that clinically responded to CRT, we did not observe a comparable increase in nonresponding patients.
There is a close association between HF and arterial blood pressure. The lower BP in patients with reduced EF leads to worse outcomes, and higher BP leads to more positive outcomes. 22  They thought that decreased arterial compliance would contribute to the pathophysiology of HF by increasing the LV wall stress. 32 Patrianakos et al evaluated arterial stiffness in patients with HFrEF.
This study compared 60 patients (mean LVEF of ~40%) with healthy controls, and found that the proximal aortic stiffness calculated by echocardiographic method was higher in the HF group. 33 Demir et al evaluated 98 patients with ischemic cardiomyopathy and found that a PWV cut-off value of 11.06 m/sec was a predictor for mortality in these patients. 6 Regnault et al investigated 306 patients in EPHESUS study, and showed that increased PWV contributed to major adverse cardiovascular outcomes. In their study, mean PWV was above 11 m/sec. 34 In our study, the mean PWV of the patients was 9.9 ± 1.5 m/sec. After CRT, it increased to 10.4 ± 1.8 m/sec (P = .009). Even though PWV increased following CRT, it is important to note that it was not above the values that were shown to be associated with poor prognosis in the literature. 35 Up to 30%-45% of patients treated with CRT do not clinically respond. 36

TA B L E 4 (Continued)
used. 38 In our study, we obtained clinical response in more than 50% of patients in accordance with the literature. Although QRS durations were wider, patients with Non-LBBB QRS morphology were less likely to respond to CRT. PWV was not significantly different in patients with and without clinical response. Therefore, our study could not suggest that PWV could be used as a predictor for response to CRT.
There is no study in the literature evaluating the post-CRT PWV change and its association with clinical response. We detected a statistically significant increase in PWV 3 months after CRT. Also we observed an increase in PWV in patients with clinical response, however, did not observe the same increase in patients without clinical response. Although increase in PWV is known to contribute to adverse clinical outcomes for cardiovascular diseases, it is not likely that CRT which was proved to increase CO, stroke volume, SBP and PP would cause a decrease in PWV. It is hard to say that the improvement of PWV independently predicts favourable CV outcome. This increase in PWV is thought to be a reflection of increased systolic blood pressure rather than an independent outcome. Further studies needed to confirm our findings. According to the data obtained in our study, modest increases in CO, SBP and PWV seen after CRT appear to be associated with positive clinical outcomes.

| S TUDY LIMITATI ON S
Several limitations have to be acknowledged. Firstly, short follow-up duration, lower statistical power and results were major limitations.