The European Society of Cardiology
Standard haemodynamic measurements
a Department of Cardiology, Heart Centre North Rhine-Westphalia, Ruhr University Bochum, Bad Oeynhausen, Germany
b Department of Thoracic and Cardiovascular Surgery, Heart Centre North Rhine-Westphalia, Ruhr University Bochum, Bad Oeynhausen, Germany
Received 3 May 2004; accepted 24 May 2004.
* Dr. Jürgen Vogt, Department of Cardiology, Heart Centre North Rhine-Westphalia, Ruhr University Bochum, Bad Oeynhausen, Germany. Tel.: +49-5731-971258; fax: +49-5731-972194
akohlstaedt{at}hdz-nrw.de
Abstract
Aims Several studies on the acute effect of cardiac resynchronization in patients with advanced heart failure (HF) and left bundle branch block (LBBB) have shown that left and biventricular stimulation increase pulse pressure and contractility, while patients with a QRS complex 
150 ms may deteriorate during stimulation.
Methods and results Patients with LBBB, severe HF and a QRS width 
150 ms underwent right, left and biventricular stimulation at different AV delays. Acute response was defined as 10% pulse pressure increase. 165 of 188 patients (88%) in sinus rhythm (47 women, mean age 62.5±10 years, ejection fraction 23±8%, NYHA class 3.1±0.3) were regarded acute responders. 10% of 103 patients with dilated cardiomyopathy and 16.5% of 79 patients with coronary artery disease were considered non-responders. 29 patients (81%) with two posterolateral veins were acute responders with 10 of them (33%) being responders in only one vein. 54 patients had an atrio-left ventricular pulse pressure increase of 10.7±10.6%, 9.8±6.4% in 48 patients with atrio-biventricular stimulation. At one-year follow-up, heart failure had significantly improved from NYHA class 3.1±0.4 to 2.1±0.7 (
), VO2peak from 12.7±2.8 to 15.9±3.6 ml/min/kg. Left ventricular enddiastolic diameter being an indicator of reverse remodelling decreased from 80.5±10.5 to 73.3±13 ().
Conclusion Haemodynamic testing before CRT allows the identification of acute non-responders as well as the best mode and site of stimulation and the optimal atrioventricular delay in responders.
Key Words: Heart failure • Resynchronization • Stimulation site • Stimulation mode • Pulse pressure • Responder
Introduction
Since 1995, numerous studies reported on the acute haemodynamic effects of cardiac resynchronization therapy (CRT). There is now convincing evidence that biventricular/left ventricular stimulation results in a decrease of pulmonary artery and wedge pressures, reduced V-wave amplitude, and increase cardiac output. The observed pulse pressure increase during acute testing correlates with increased stroke volume, and the rise in left ventricular 
representing improved contractility. After permanent implantation of CRT devices, pressure-volume loop analysis demonstrated differences between left/biventricular and right ventricular pacing, an effect independent from the respective right ventricular (RV) pacing site (RV outflow tract versus RV apex). In contrast to RV pacing, left/biventricular stimulation results in a decrease of end-systolic volume and an increase in stroke volume. In contrast to bi- and left ventricular pacing, various studies proved that right ventricular stimulation is associated with only marginal benefits, even with optimized AV delays (Table 1).1–11
|
In the PATH-CHF-I trial, the impact of QRS complex widths greater than 150 ms on acute efficacy was first demonstrated.8 Compared to right ventricular stimulation, patients with QRS complexes wider than 150 ms had significantly higher pulse-pressures and
values with either left or biventricular pacing and shortened AV intervals. This contrasts to worsening left ventricular function, reduced contractility and pulse pressures observed in some patients with narrow complexes (150 ms) when the AV interval was shortened.
Nelson et al. elucidated that in their patient population no single parameter, not even QRS complex width, was predictive for the CRT result, while the combination of a QRS complex width 155 ms and markedly reduced contractility indicated by a below 700 mmHg/s was a strong predictor for the pulse pressure to increase by at best 10% and contractility by 25%.9
Standard acute testing
Acute haemodynamic determination of left and right ventricular contractility (), aortic pulse pressure, and particularly pressure-volume loop measurements require the insertion of multiple catheters, a complex procedure particularly invasive and stressful especially for severely ill and haemodynamically compromised patients with advanced heart failure.7 To improve the feasibility and practicability of acute haemodynamic testing, we investigated the role of arterial pulse pressure as the only parameter. Based on our empiric findings in patients with severe heart failure and optimal pharmacological management having an average pulse pressure of 35–45 mmHg, we considered a pulse pressure increase by 
10% above baseline as significant and indicative for a positive response to the respective pacing mode.
Patients were examined under local anaesthesia in the cardiac catheterization laboratory after having been titrated to their optimal heart failure medication. In addition to installing a 12-lead ECG for QRS width monitoring, bipolar electrophysiologic 5F pacing and sensing catheters were advanced into stable positions in the right ventricular apex, in some cases the right ventricular outflow tract, the right atrial appendage or the superior right atrium, if patients had previous open-heart surgery. The pulse pressure was recorded invasively from a 4F introducer in the femoral artery.
The coronary sinus was cannulated with an 8F Amplatz guiding catheter (custom-made, 60 cm long), mostly using a size 2 catheter, and size 1 for smaller atria. Following injection of contrast medium via the guiding catheter, a 6F Swan-Ganz catheter was advanced into the proximal, middle, or more distal section of the coronary sinus, in some cases under guidewire assistance. After balloon inflation, the coronary sinus and its tributaries were visualized in two angiographic planes (LAO and PA).
In accordance with the respective venous anatomy and in accordance with the publication by Butter et al.,11 only posterior, posterolateral, and higher posterolateral (lateral) veins were accepted as target vessels and cannulated. For stimulation, a quadripolar 0.018
pacing wire (Pathfinder, Cardima) was advanced into the distal third of the target vein, either through the Amplatz catheter, the Swan-Ganz catheter or via a special 5F angiographic catheter selected to match the individual anatomy.
After determining the baseline resting pulse pressure and heart rate, atrially triggered pacing commenced with an external DDD pacemaker equipped with a custom-made Y-adapter connected to both the right ventricular and left ventricular electrophysiologic catheter.
To prevent offsetting of the resting pressure, steady-state pacing was not performed. Ten beats each were paced in atrio-leftventricular, atrio-biventricular, and atrio-rightventricular mode, with AV intervals of 220, 200, 180, 150, 120, 100, and 80 ms each, interrupted by 30 s without pacing. For each AV interval step, pulse pressure was calculated by subtracting the diastolic pressure from the systolic pressure, both averaged over 10 beats. Resting pulse pressure and heart rate were again recorded between the three pacing modes, in order to eliminate the effect of pressure fluctuations. With each step, the QRS width was simultaneously determined from a 12-lead ECG obtained with an electrophysiology rack. Any pulse pressure increase of 10% or more was considered a significant acute response (Fig. 1).
|
Prior to pacing, the RV–LV signal delay was measured between right ventricular endocardium and coronary target vein (Fig. 2).
|
Patient population
We examined 188 consecutive patients, who had consented to undergo acute haemodynamic testing. All patients were in regular sinus rhythm. 47 patients were female; mean patient age was 62.5±10 years. The ejection fraction of these patients with advanced heart failure was 23±8%. Correspondingly, the mean NYHA class was 3.1±0.3. In accordance with the selection criteria regarding QRS width, the mean QRS duration was 199±22 ms, and the mean PQ duration was 220±40 ms. Mean peak exercise oxygen consumption as determined by cardiopulmonary exercise (CPX) testing was 13.1±30 ml/m2/min prior to aborting CPX. Before and while undergoing acute measurements, all patients received optimal medical heart failure treatment.
Results
An acute haemodynamic response was observed in 165 out of 188 patients (88%). No acute response as defined by a pulse pressure increase of less than 10% was documented in 23 patients (12%). Among 103 patients with dilated cardiomyopathy, 10 patients (10%) were non-responders, while 13 (16.5%) of 79 patients with ischaemic heart disease failed to demonstrate a significant pulse pressure increase. Six patients (all of them responders) had LV pump failure due to chronic valvular disease.
Some parameters related to the AV delay were significantly different between responders and non-responders (Table 2). QRS width was 193±21 ms in responders, i.e. significantly longer than in non-responders (175±21 ms). The delay between right and left ventricle (RV–LV delay) was also significantly longer in responders (114±29 ms) than in non-responders (91±29 ms).
|
The RV–LV delay in responders with underlying dilated cardiomyopathy was 120±27 ms, which was significantly longer than that of non-responders (94±29 ms). Different structural pathologic changes due to transmural myocardial scarring or diffuse intramyocardial fibrosis are probably the reason why the RV–LV delay was also significantly longer in acute responders with dilated cardiomyopathy (120±27 ms) than in acute responders with ischaemic heart disease (104±29 ms). Due to overlap, there was no clear correlation between RV–LV delay and acute response.
Pulse pressure improvement with different posterolateral pacing sites
In the literature it has already been elaborated how pacing the posterolateral free wall of the left ventricle affects pulse pressure and differently than pacing the anterior or anterolateral regions. Butter et al.11 were able to show that in all patients, pulse pressure and were markedly lower with anterolateral or anterior compared to posterolateral pacing. The fact that some patients experience haemodynamic deterioration, including decreasing contractility and pulse pressure, if paced from the region of the anterior vein, has also been addressed. This effect does not depend on the specific AV delay chosen. We aimed to demonstrate possible differences between pacing from various posterolateral free wall sites utilizing different posterolateral veins. In 36 patients, pacing commenced from two or more posterolateral veins; 81% (29 patients) were responders. With respect to pulse pressure increase, the mean difference between two venous sites was 7±6.7%. In 10 of 29 patients, a significant pulse pressure increase was only observed when stimulating from one of the alternate venous sites.
Optimal pacing mode
From the literature it has not yet conclusively been shown that the haemodynamic effects of atrio-leftventricular versus atrio-biventricular pacing are different. In 37% of the responders (59 patients), the pulse pressure difference between left and biventricular stimulation was 3% or less, whereas atrio-leftventricular pacing resulted in improved haemodynamics in 54 patients whose pulse pressure increased by 10.7±10.6%. Biventricular pacing was superior in 48 patients, resulting in an average pulse pressure increase of 9.8±6.4%. Two additional patients who had neither responded to left ventricular nor to atrio-biventricular pacing, including pacing from the right ventricular apex, experienced a significant pulse pressure rise under biventricular pacing after the right ventricular pacing lead had been moved to the right ventricular outflow tract. In 44 patients, a significant haemodynamic acute response was only achieved with one pacing mode; 19 patients did not respond to biventricular and 21 patients did not respond to left ventricular pacing (Fig. 3).
|
Optimal AV delay
When analyzing the best AV delay resulting in the highest pulse pressure increase with either atrio-leftventricular or atrio-biventricular stimulation being the optimal pacing mode, the best compromise between intrinsic septal and left ventricular free wall excitation was achieved with a relatively long AV delay for both stimulation modes. The best AV delay for atrio-leftventricular pacing was 155.2±40.6 ms, which is equivalent to 71% of the intrinsic PQ interval; the pulse pressure increased by 26±17%. With atrio-biventricular stimulation, the best AV delay was 161.3±37.4 ms, i.e. 73% of the intrinsic PQ interval, resulting in an average pulse pressure increase of 28±18%.
In 80% of patients with intrinsic PQ intervals longer than 190 ms, the optimal AV delay was in the range of 150 to greater than 200 ms. In 84% of patients with an intrinsic PQ interval of 190 ms, the optimal AV delay was in the range of 120 to 180 ms, in 50% it was 150 ms.
Long-term follow-up
In all patients who exhibited an acute response, a left ventricular lead was implanted into the target vein found most suitable during acute testing, whenever technically feasible. The best pacing mode was activated. Weighing the unproven clinical benefit against the complicated and stressful procedure, non-responders did not receive implants. During the follow-up period of 14±10 months, acute responders with implants experienced a highly significant clinical improvement under optimally tailored treatment: the NYHA class improved from 3.1±0.4 to 2.1±0.7, peak oxygen consumption under exercise increased from 12.7±2.8 to 15.9±3.6ml/min/kg (). The left ventricular enddiastolic diameter decreased from 80.5±10.5 to 73.3±13 mm (
) (Table 3) indicating left ventricular reverse remodelling.
|
Discussion
A large group of patients with advanced heart failure, sinus rhythm, and a QRS duration of 150 ms, who were potential candidates for cardiac resynchronization therapy, underwent acute haemodynamic testing. 12% of the entire study group were non-responders. While only 10% of patients with underlying dilated cardiomyopathy had an insignificant pulse pressure increase, 16% of patients with coronary heart disease did not respond. This observation illustrates how the underlying pathology may affect the success of cardiac resynchronization. In most large resynchronization trials, biventricular stimulation was uniformly chosen as pacing mode.12 Occasionally, technical reasons prevented pacing from the optimal left ventricular free wall region and other sites were used. Some of these studies also included patients with narrower QRS complexes. Even among these large groups, up to 30% of patients did not experience long-term benefits.13 Standard acute testing in our patient group demonstrated that 33% of the responders had no or a smaller pulse pressure rise under biventricular than under left ventricular stimulation. The clear-cut difference in response to bi- versus left ventricular pacing in some patients illustrates that just the pacing site chosen may decide whether a patient turns out to be a responder or a non-responder.
One needs to concede that in the past pure left ventricular stimulation had not always been possible, since a higher amplitude safety margin with a pseudo-bipolar pacing setup created more a biventricular stimulation pattern than the left ventricular stimulation intended. Precise bipolar pacing has become possible since bipolar coronary vein leads became available. Alternate venous pacing sites at the posterolateral free wall are associated with specific intraventricular conduction delays, resulting in distinct responses when various regions are stimulated. In this context it was particularly interesting to observe that one third of our patients in whom several posterolateral target veins were tested, responded only to pacing from one of these vessels.
The long-term follow-up data in this patient group are convincing with respect to the most important prognostic parameters: NYHA heart failure class, increased peak exercise oxygen consumption, and a decrease of left ventricular dimension indicating reversed remodelling. Whereas large trials on cardiac resynchronization reported maximal VO2peak increases of 1.5 ml/kg/min, VO2peak in our population increased by 3.2 ml/kg/min after 12 months.
Preoperative haemodynamic testing can immediately distinguish non-responders from responders. In responders, testing will identify the most efficient pacing mode, the best pacing site, and the optimum atrio-ventricular delay, i.e. the optimal fusion of intrinsic septal excitation and paced excitation of the left ventricular free wall. Long-term follow-up suggest superior results if cardiac resynchronization is individually tailored.
The mean RV–LV delay can be used to distinguish responders from non-responders, but too much overlap prevents the definition of a fixed cut-off point predicting the acute and long-term response.
Correlating hemodynamic data with derived echocardiographic dyssynchrony parameters is expected to streamline pre-implant testing and to permit more precise tailoring of resynchronization management.
Footnotes
1 Both authors made equal contributions to this study. 
References
- Gold MR, Feliciano Z, Gottlieb SS, et al. Dual-chamber pacing with a short atrioventricular delay in congestive heart failure: a randomized study. J. Am. Coll. Cardiol. 1995;26:967–973[Abstract]
- Foster AH, Gold MR, MacLaughlin JS. Acute hemodynamic effects of atrio-biventricular pacing in humans. Ann. Thorac. Surg. 1995;59:294–300
[Abstract/Free Full Text] - Cazaeu S, Ritter P, Lazarus A, et al. Multisite pacing for end-stage heart failure: early experience. PACE. 1996;19:1748–1757
- Blanc JJ, Etienne Y, Gilard M, et al. Evaluation of different ventricular pacing sites in patients with severe heart failure: results of an acute hemodynamic study. Circulation. 1997;96:3273–3277
[Abstract/Free Full Text] - Saxon LA, Kerwin WF, Cahalan MK, et al. Acute effects of intraoperative multisite ventricular pacing on left ventricular function and activation/contraction sequence in patients with depressed ventricular function. J. Cardiovasc. Electrophysiol. 1998;9:13–21[Web of Science][Medline]
- Leclercq C, Cazeau S, Le Breton H, et al. Acute hemodynamic effects of biventricular DDD pacing in patients with end-stage heart failure. J. Am. Coll. Cardiol. 1998;32:1825–1831
[Abstract/Free Full Text] - Kass DA, Chen CH, Curry C, et al. Improved left ventricular mechanics from acute VD pacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circulation. 1999;99:1567–1573
[Abstract/Free Full Text] - Auricchio A, Stellbrink C, Block M, et al. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. Circulation. 1999;99:2993–3001
[Abstract/Free Full Text] - Nelson GS, Curry CW, Wyman BT, et al. Predictors of systolic augmentation from left ventricular preexcitation in patients with dilated cardiomyopathy and intraventricular conduction delay. Circulation. 2000;101:2703–2709
[Abstract/Free Full Text] - Nelson GS, Berger RD, Fetics BJ, et al. Left ventricular or biventricular pacing improves cardiac function at diminished energy cost in patients with dilated cardiomyopathy and left bundle-branch block. Circulation. 2000;102:3053–3059
[Abstract/Free Full Text] - Butter C, Auricchio A, Stellbrink C, et al. Effect of resynchronization therapy stimulation site on the systolic function of heart failure patients. Circulation. 2001;104:3026–3029
[Abstract/Free Full Text] - The CONTAK CD Investigators: cardiac resynchronization therapy in heart failure (CONTAK CD) clinical study. Presented at the meeting of the North American Society of Pacing Electrophysiologists, Boston, MA; 2001
- Abraham WT, Fisher WG, Smith AL, et al. Cardiac resynchronization in chronic heart failure. N. Engl. J. Med. 2002;346:1845–1853
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  O. Oldenburg, L. Faber, J. Vogt, A. Dorszewski, F. Szabados, D. Horstkotte, and B. Lamp Influence of cardiac resynchronisation therapy on different types of sleep disordered breathing Eur J Heart Fail, August 1, 2007; 9(8): 820 - 826. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||