Author + information
- Received December 22, 2006
- Revision received April 10, 2007
- Accepted April 16, 2007
- Published online August 21, 2007.
- Cheuk-Man Yu, MD, FRCP, FRACP⁎,⁎ (, )
- Fang Fang, MM, PhD†,
- Qing Zhang, MM, PhD⁎,
- Gabriel W.K. Yip, MD⁎,
- Chun Mei Li, BM⁎,
- Joseph Yat-Sun Chan, FHKAM⁎,
- LiWen Wu, BM⁎ and
- Jeffrey Wing-Hong Fung, FRCP⁎
- ↵⁎Reprint requests and correspondence:
Prof. Cheuk-Man Yu, Li Ka Shing Institute of Health Sciences, Institute of Vascular Medicine, S. H. Ho Cardiovascular and Stroke Centre, Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong.
Objectives We sought to examine whether cardiac resynchronization therapy (CRT) improves atrial function and induces atrial reverse remodeling.
Background Cardiac resynchronization therapy is an established therapy for advanced heart failure with prolonged QRS duration, which improves left ventricle (LV) function and is associated with LV reverse remodeling.
Methods A total of 107 heart failure patients (66 ± 11 years) who received CRT and were followed up for 3 months were studied. Atrial function was assessed by M-mode, 2-dimensional echocardiography, transmitral Doppler, tissue Doppler velocity, and strain (ε) imaging. Left atrial (LA) emptying fraction based on the change in areas (LAA-EF) and volumes (LAV-EF) were calculated. The LV reverse remodeling was defined by a reduction of LV end-systolic volume >10%.
Results In the responders of LV reverse remodeling (n = 62), LAA-EF and LAV-EF were significantly increased (p < 0.001). Responders also had significant decrease in LA size area and volumetric measurements, both before (p < 0.05) and after atrial systole (p < 0.001). However, these parameters were unchanged in the nonresponders (n = 45, p = NS). In the responders, tissue Doppler velocity analysis showed improvement of contraction velocity in both left (p = 0.005) and right atria (p = 0.018), whereas ε in both atria were increased in all the phases of cardiac cycle, namely ventricular end-systole (p < 0.001), early diastole (p < 0.001), and late diastole (p = 0.007).
Conclusions Cardiac resynchronization therapy improves both left and right atrial pump function. The increase in atrial ε throughout the cardiac cycle is likely reflecting the improvement of atrial compliance. These changes lead to LA reverse remodeling with reduction of LA size before and after atrial systole.
Cardiac resynchronization therapy (CRT) is now an established treatment for patients with advanced heart failure with prolonged QRS duration. Apart from clinical benefits, improvement of left ventricular (LV) systolic function and associated LV reverse remodeling have been well reported (1–6). Recently, improvement of right ventricular function also has been reported (7). Despite the extensive evidence of the benefits of CRT on ventricular function, whether the use of CRT benefits patients with atrial function has not been evaluated. With improvement of LV function and reduction of mitral regurgitation, left atrial (LA) size could be reduced.
Furthermore, the pressure unloading effect in the atrium may result in the improvement of atrial function. Atrial function is relatively complex. Apart from active atrial pump function as a direct result of atrial systole, atrial compliance is an important determinant of atrial reservoir and conduit functions (8). With the advancement of echocardiographic technology, it is now possible to assess regional atrial function, in particular by tissue Doppler velocity and strain (ε) imaging. Tissue Doppler velocity is useful to assess regional atrial active contractile function, whereas tissue Doppler strain is a good measure of myocardial deformation (9,10). Therefore, in the present study we combined the use of conventional and advanced echocardiographic tools of tissue Doppler velocity and strain to examine atrial function and determined whether atrial reverse remodeling occurred and atrial function improved after CRT. Furthermore, whether such changes were different between responders and non-responders of LV reverse remodeling were also explored.
The study population consisted of 120 consecutive patients with advanced congestive heart failure who had received CRT. The inclusion criteria of CRT included symptomatic heart failure despite optimal pharmacological therapy, New York Heart Association (NYHA) functional class III or IV heart failure, ejection fraction <40% and QRS duration >120 ms in the form of bundle branch block or intraventricular conduction delay. They were followed up at 3 months after the therapy, at which point 3 patients had died and 1 patient had dropped out before follow-up. Serial standard echocardiography with tissue Doppler imaging and clinical assessment were performed at baseline and 3 months after CRT. Nine patients whose major echocardiographic parameters could not be obtained because of poor image quality were excluded from the study. As a result, 107 patients (66 ± 11 years, 75% males) were included in the analysis. Among them, 11 patients had permanent atrial fibrillation, but the ablation of atrioventricular node was only attempted in 3 patients with uncontrolled ventricular rate. The study protocol was approved by the Ethics Committee of The Chinese University of Hong Kong and written informed consent was obtained from each participant.
Biventricular device implantation
Biventricular devices were implanted as previously described (4,6). The LV pacing lead was inserted by a transvenous approach through the coronary sinus to target lateral or posterolateral cardiac vein. Choices of CRT devices included biventricular pacemaker in 97 patients (InSync, InSync III from Medtronic Inc., Minneapolis, Minnesota; Contak TR or Contak TR II from Guidant Inc., St. Paul, Minnesota) and biventricular defibrillator in 10 patients (InSync ICD, InSync Marquis or InSync Sentry from Medtronic Inc., Minneapolis, Minnesota; Contak CD or Contak Renewal from Guidant Inc., St. Paul, Minnesota).
Echocardiographic assessment of atrial size and function
Echocardiography with tissue Doppler imaging was performed (Vivid 5 or Vivid 7, Vingmed-General Electric, Horten, Norway) serially before and 3 months after CRT. The atrioventricular interval was optimized by Ritter’s method at day 1 after implantation to reach maximal transmitral diastolic filling and maximal biventricular capture. The adjustment of interventricular interval was not performed and, therefore, all patients were having simultaneous biventricular pacing at the default setting. The LV volumes and ejection fraction were assessed by biplane Simpson’s equation using the apical 4-and 2-chamber views where the length of the ventricular image was maximized. Patients who had a reduction of LV end-systolic volume of >10% were defined as volumetric responders of CRT, whereas those with a lesser degree of reduction of ≤10% were called nonresponders (11). Diastolic dysfunction was graded as abnormal relaxation, pseudonormal, and restrictive filling patterns, as previously described (12).
Atrial function was assessed at apical 4-chamber (4ch) and 3-chamber views (3ch) (13). In the LA, the long-axis diameter at end-diastole was measured at 4-ch view. The LA areas were then measured at 4ch and 3ch views at the following phases of the cardiac cycle: the maximal LA areas at ventricular end-systole where LA size is maximal (LAA-max-4ch and -3ch), LA areas just before atrial systole (LAA-pre-4ch and -3ch), and the minimal LA areas after atrial systole (LAA-post-4ch and -3ch). Atrial emptying fraction was calculated based on the change of areas before and after atrial systole (LAA-EF-4ch and -3ch). Using the modified Simpson rule, the atrial volume at ventricular end-systole, just before and after atrial systole and atrial emptying fraction (LAV-EF) also were calculated (13).
Tissue Doppler imaging was performed at the apical 4ch view for the long-axis motion of the heart as previously described (14,15). Two-dimensional echocardiography with color tissue Doppler imaging was performed. The imaging angle was adjusted to ensure a parallel alignment of the sampling window with the myocardial segment of interest. Gain settings, filters, pulse repetitive frequency, sector size, and depth were adjusted to optimize color saturation. At least 3 consecutive beats were stored, and the images were digitized and analyzed off-line with EchoPac-PC 6.0.1 (Vingmed-General Electric). Atrial Doppler velocity profile signals were reconstituted off-line by placing a 3 × 12-mm sampling window at the mid levels of LA, interatrial septum (IAS), and RA, respectively. The peak regional atrial contraction velocities at atrial systole (after the onset of P wave of electrocardiogram) were measured. Atrial ε was measured in the same atrial locations and was calculated by the formula: ε = (L − L0)/L0× 100%, in which L denotes the instantaneous length, and L0denotes the original length. The following parameters of atrial function by tissue Doppler velocity and ε imaging were measured:
• Left atrial contraction velocity during atrial systole;
• Inter-atrial septum contraction velocity during atrial systole;
• Right atrial contraction velocity during atrial systole;
• Left atrial ε during ventricular end-systole;
• Interatrial septum ε during ventricular end-systole;
• Right atrial ε during ventricular end-systole;
• Left atrial ε during ventricular early diastole;
• Inter-atrial septum ε during ventricular early diastole;
• Right atrial ε during ventricular early diastole;
• Left atrial ε during ventricular late diastole;
• Interatrial septum ε during ventricular late diastole; and
• Right atrial ε during ventricular late diastole
The intraobserver and interobserver variability for atrial myocardial velocity measurement were 3.2% and 4.7%, and for atrial strain measurement, they were 7.1% and 8.4%, respectively.
For comparison of continuous parametric variables between baseline and 3 months after CRT, a paired sample ttest was used. The nonparametric Wilcoxon test was adopted for comparison of ordinal variables, including NYHA functional class and pattern of diastolic dysfunction. The comparison of echocardiographic parameters between volumetric responders and nonresponders was performed with the unpaired ttest. All parametric data were expressed as mean ± SD. A p value <0.05 was considered statistically significant.
The optimal atrioventricular delay programmed was 100 ± 25 ms. Medications for heart failure were kept unchanged throughout the study period, except intravenous diuretics with/without subsequent increase in dosage of oral diuretics in case of acute decompensation. Heart rate was decreased slightly at 3 months when compared with baseline (67 ± 12 beats/min vs. 70 ± 16 beats/min, p = 0.02).
Clinical status and ventricular function
There was a favorable improvement of clinical status in the whole group, namely NYHA functional class, Minnesota Living With Heart Failure Quality of Life score, and 6-Minute Hall-Walk distance after CRT for 3 months (all p < 0.001) (Table 1).Left ventricular function was improved with the evidence of LV reverse remodeling, increase in sphericity indices, reduction of mitral regurgitation, decrease in myocardial performance index, as well as increase in diastolic filling time and favorable change in LV diastolic filling pattern (all p < 0.001) (Table 1).
Atrial remodeling and atrial function after CRT
The use of CRT resulted in significant reduction of LA area before and after atrial contraction at both 4ch and 3ch views, resulting in a significant increase in LA emptying fraction (i.e., LAA-EF-4ch and -3ch; both p < 0.001) (Table 2).Similarly, the LA emptying fraction by volumetric measurement, i.e., LAV-EF, was increased significantly (p < 0.001). However, the maximal LA area and volume at ventricular end-systole were unchanged. There also was no change in LA diameter observed. Transmitral Doppler measurement of peak atrial velocity showed no difference between baseline and 3 months after CRT.
Left and right atrial function was further examined by tissue Doppler velocity and strain imaging. It was observed that atrial contraction velocities in all the 3 atrial sites were improved significantly (Table 2). Atrial strain at ventricular end-systole, after early-diastole, as well as at end-diastole, were improved.
Atrial remodeling, atrial function, and CRT response
There were 62 responders (58%) and 45 (42%) nonresponders of LV reverse remodeling. There were 5 patients with permanent AF in the nonresponder group and 6 in the responder group. The changes in atrial structure and function were divergent in the 2 groups (Table 3).In the nonresponders, there was no evidence of reduction in LA area or volume in all the 3 phases of the cardiac cycle (Fig. 1).As a result, no improvement of atrial emptying fraction was observed (both calculated by area or by volume), in contrast to the responders that the LA area and volume before and after atrial contractions were significantly reduced (Table 3, Fig. 1). As a result, LA emptying fraction by LAA-EF-4ch, LAA-EF-3ch, and LAV-EF were significantly improved (all p < 0.001).
By tissue Doppler velocity and strain imaging, it was observed that improvement of LA contraction velocity was only observed in the responders, though RA and IAS contraction velocity was improved in both groups (Table 3, Fig. 2).Furthermore, improvement of LA and RA strain during ventricular end-systole and early diastole were only observed in the responders (Table 3, Fig. 3).
There was no significant change in transmitral peak atrial velocity in both responders and non-responders of LV reverse remodeling. There was reduction in mitral regurgitation in both responders (32 ± 19% vs. 22 ± 19%, p < 0.001) and nonresponders (34 ± 22% vs. 25 ± 19%, p = 0.005) of reverse remodeling, though the magnitude was similar in both groups (−10 ± 13% vs. −8 ± 18%, p = NS). Furthermore, when those patients with atrial fibrillation were excluded and the analyses were repeated, the above observations remained unchanged.
This study examined atrial function and atrial remodeling in heart failure patients who received CRT by the combined use of conventional and new echocardiographic imaging tools. It was observed that active atrial contractile function was significantly improved in both atria, in particular the LA. Furthermore, atrial remodeling is evident by the reduction of atrial area and volume before and after atrial systole. These findings corroborated with the improvement of atrial function, including the increase of atrial emptying fraction, peak atrial contraction velocity by tissue Doppler velocity, and atrial ε. There was also improvement of atrial ε during ventricular end-systole and early diastole, which may suggest the improvement of atrial compliance. Of note, these changes were mainly observed in responders of LV reverse remodeling.
Improvement of atrial function and atrial reverse remodeling after CRT
Atrial function is an integral part of cardiac function, as at least one-third or more of the LV filling is dependent on active atrial pump function, especially in the elderly population (8). In heart failure patients who received CRT, a number of studies have confirmed the improvement of LV systolic function, LV reverse remodeling, as well as reduction of LV mass (5,6,11,16). However, atrial function and atrial reverse remodeling have not been explored. This study examined atrial function by the combined use of conventional and new echocardiographic technologies. It appeared that LA active contractile function is improved in the responders of LV reverse remodeling after CRT, as suggested by the increase in LA emptying fraction by all the 3 methods. Furthermore, tissue Doppler imaging revealed the increase in atrial contraction velocity and atrial ε during ventricular end-diastole. In fact, the atrial contraction velocities by tissue Doppler imaging were also increased in IAS and RA, which reflect the improvement of biatrial contractile function after CRT.
Our analysis also observed the impact of LV reverse remodeling response on atrial size and function. In fact, the favorable changes in LA size and contractile function were only observed in responders of LV reverse remodeling but not the nonresponders. As shown in previous studies, responders of LV reverse remodeling had evidence of more efficient diastolic filling (4,5), favorable change in geometry (high sphericity index that reduced LV wall tension) (16), reduction of mitral regurgitation (17,18), lower LV diastolic filling pressure (19), as well as regression of LV mass (6). This will favor pressure and volume-unloading effects in the LA, which facilitate atrial emptying. Furthermore, a recent study also suggested the favorable improvement of RV function and reduction of tricuspid regurgitation in CRT responders, which may lead to favorable change in RA function (7). As the magnitude of reduction in mitral regurgitation after CRT was similar between the 2 groups, reduction volume overload caused by functional mitral regurgitation is unlikely to be the major contributing factor for atrial reverse remodeling after CRT.
Improvement of atrial strain after CRT and its implication
It is also important to note that the maximal LA area and volume at end-systole remain unchanged despite the improvement of active atrial contractile function. Nevertheless, left and right atrial ε were increased during ventricular systole and early diastole. Strain is a measure of regional deformity, which is dependent on the ultrastructural components of the atrium, such as the extent of atrial myocyte hypertrophy and amount of interstitial fibrosis (20). Therefore, our results suggest that the atria are more elastic and compliant when subjected to passive stretching. In other words, this reflects improvement of both reservoir (increased atrial ε during ventricle systole) and conduit functions (increased atrial ε during early diastole). Therefore, although the maximal atrial size remained unchanged in the responders, the more dynamic changes in atrial ε throughout the cardiac cycle may possibly reflect structural changes leading to better atrial compliance.
Regarding the possible mechanism of beneficial changes in atrial function after CRT, it will be more comprehensive if factors other than LV reverse remodeling were also taken into consideration, such as assessment of severity and change in mitral regurgitation. However, the present study aimed at investigating the changes of atrial remodeling and function as the primary objective. Furthermore, more quantitative methods are needed to explore the relationship between the changes of mitral regurgitation and LA functional improvement, such as the use of proximal isovelocity surface area method, which might not be applicable in some patients with less than mild-to-moderate or eccentric mitral regurgitation. It will also be interesting to explore whether atrial reverse remodeling is associated with reduction of atrial arrhythmias, which was not examined in the present study. Finally, there were a number of comparisons of echocardiographic parameters between baseline and 3 months, as well as between responders and nonresponders, which may have added risk of type I error.
Our study illustrated that the use of CRT improved LA and RA contractile function. The increase of atrial ε throughout the cardiac cycle is likely reflecting the improvement of atrial compliance. These changes lead to LA reverse remodeling with reduction of LA size before and after atrial systole.
This study was supported by a research grant from Li Ka Shing Institute of Health Sciences.
- Abbreviations and Acronyms
- cardiac resynchronization therapy
- interatrial septum
- left atrial
- left atrial emptying fraction based on the change in areas
- left atrial emptying fraction based on the change in volumes
- left ventricular
- New York Heart Association
- right atrial
- Received December 22, 2006.
- Revision received April 10, 2007.
- Accepted April 16, 2007.
- American College of Cardiology Foundation
- Yu C.M.,
- Chau E.,
- Sanderson J.E.,
- et al.
- St John Sutton M.G.,
- Plappert T.,
- Abraham W.T.,
- et al.
- Zhang Q.,
- Fung J.W.,
- Auricchio A.,
- et al.
- Bleeker G.B.,
- Schalij M.J.,
- Nihoyannopoulos P.,
- et al.
- Stefanadis C.,
- Dernellis J.,
- Toutouzas P.
- Yu C.M.,
- Bleeker G.B.,
- Fung J.W.,
- et al.
- Yu C.M.,
- Sanderson J.E.,
- Shum I.O.,
- et al.
- Yu C.M.,
- Wang Q.,
- Lau C.P.,
- et al.
- Yu C.M.,
- Fung J.W.,
- Zhang Q.,
- et al.
- Breithardt O.A.,
- Sinha A.M.,
- Schwammenthal E.,
- et al.
- Kanzaki H.,
- Bazaz R.,
- Schwartzman D.,
- Dohi K.,
- Sade L.E.,
- Gorcsan J. III.
- Ohtani K.,
- Yutani C.,
- Nagata S.,
- Koretsune Y.,
- Hori M.,
- Kamada T.