- Email: [email protected]
Short-spaced dipole for managing phrenic nerve stimulation in patients with CRT: The “phrenic nerve mapping and stimulation EP” catheter study
1
Short-spaced dipole for managing phrenic nerve stimulation in patients with CRT: The “phrenic nerve mapping and stimulation EP” catheter study Mauro Biffi, MD,* Francesco Zanon, MD,† Emanuele Bertaglia, MD,‡ Luigi Padeletti, MD, PhD,y Annamaria Varbaro, MS,k Tiziana De Santo, MSc,k Giuseppe Boriani, MD, PhD,* Zhongping Yang, PhDz on behalf of the PING-EP Study Group From the *Institute of Cardiology, University of Bologna, Bologna, Italy, yOspedale Santa Maria della Misericordia, Rovigo, Italy, zOspedale Civile, Mirano, Italy, yUniversity of Florence, Florence, Italy, kMedtronic Italia SpA, Milano, Italy and z Medtronic Inc, Minneapolis, Minnesota. BACKGROUND Phrenic nerve stimulation (PNS), occurring in 33%–37% of the patients with cardiac resynchronization therapy (CRT), is a limiting factor when implanting left ventricular (LV) leads from coronary veins. OBJECTIVE To test the hypothesis that PNS occurence is related to bipolar electrode spacing. METHODS During standard CRT defibrillator implant procedures, a 5-F diagnostic electrophysiology catheter with 10 electrodes, spaced 2–5–2 mm, was positioned in a cardiac vein suitable for permanent LV lead placement. Pacing in the unipolar configuration identified the site with the lowest PNS threshold. PNS and left ventricular pacing (LVP) thresholds were then measured in different configurations at 0.5 ms: unipolar, each LV electrode served as the cathode in turn; and bipolar with different electrode spacing, cathode being the electrode with the lowest unipolar PNS threshold. RESULTS From February to September 2010, 40 patients undergoing CRT implantation were enrolled in 4 centers in Italy. It was possible to identify PNS and perform a complete set of
measurements in 23 patients. A bipolar electrode spacing of 2 mm resulted in higher PNS thresholds in bipolar configurations han did a bipolar electrode spacing of Z5 mm. However, no significant increase in the LVP threshold was observed (P ¼ ns). CONCLUSIONS This experience suggests that LVP with a bipolar electrode spacing of 2 mm significantly increases the PNS threshold without affecting the LVP threshold, thereby increasing the possibility of delivering CRT when the LV lead is placed in proximity to the phrenic nerve. KEYWORDS Cardiac resynchronization therapy; Phrenic nerve stimulation; Bipolar electrode spacing; Threshold ABBREVIATIONS CRT ¼ cardiac resynchronization therapy; EP ¼ electrophysiology; LV ¼ left ventricular; LVP ¼ left ventricular pacing; PNS ¼ phrenic nerve stimulation; RV ¼ right ventricular; TPNE ¼ targeted phrenic nerve electrode (Heart Rhythm 2013;10:39–45) I 2013 The Heart Rhythm Society. All rights reserved.
3 3
Introduction Cardiac resynchronization therapy (CRT) is an established treatment of heart failure caused by left ventricular (LV) systolic dysfunction with evidence of electrical/mechanical dyssynchrony.1,2 CRT aims at improving the LV mechanics by stimulating the mostly delayed LV sites.3–5 This approach has been shown to result in clinical benefits.5–7 Phrenic nerve stimulation (PNS) is a challenge to delivering effective CRT. It has been observed in 33%–37% of the patients8–10 and is more common at the same LV sites where the chance of a Dr Biffi received a modest honorarium as speaker’s bureau and education activity. Dr Yang, Ms Varbaro, and Ms Santo are employees of Medtronic. Address reprint requests and correspondence: Dr Mauro Biffi, MD, Institute of Cardiology, University of Bologna, Via Massarenti 9, 40138 Bologna, Italy. E-mail address: [email protected].
favorable response to CRT is maximal.5,8 Although PNS is actively addressed at implantation,8 the long-term management of PNS may present a challenge in patients with CRT.11–13 Indeed, about 15% of the patients need to be reevaluated because of PNS occurrence at follow-up11–13 and about 6.6% eventually report long-term PNS symptoms despite repeated attempts at PNS avoidance.12 Moreover, the absence of PNS at implantation does not ensure freedom from PNS at follow-up.8–10 Although built-in features, for example, automated LV output adjustment and capability to separately program LV lead cathode/anode can help manage PNS,8,10,11,14 a definitive solution to this challenge is not available.15 We sought to investigate the physiologic principles of PNS in an acute study of CRT candidates in order to develop a comprehensive strategy aimed at avoiding PNS symptoms in clinical practice.
1547-5271/$-see front matter B 2013 The Heart Rhythm Society. All rights reserved.
http://dx.doi.org/10.1016/j.hrthm.2012.08.045
40
Heart Rhythm, Vol 10, No 1, January 2013
Figure 1 Fluoro image of coronary vein angiogram and of electrophysiology catheter placed into the coronary vein. Lead tip is the targeted phrenic nerve electrode (TPNE).
Methods We investigated the relationship between the distance of the left ventricular pacing (LVP) cathode from the phrenic nerve and the PNS threshold in the unipolar configuration (LV cathode to RV coil) as well as the relationship between bipolar electrode spacing and the PNS threshold in the true bipolar configuration (LV cathode to LV anode) when the LV cathode was placed at the site with the lowest PNS threshold. This study was carried out at 4 Italian hospitals with more than 8 years of experience in CRT implantation. The study protocol was approved by the hospital ethical committee of each center. During February to September 2010, patients scheduled for a CRT implantation were asked to participate in this clinical study by signing an informed consent: in the event PNS was detected during LV lead implantation, a detailed study of PNS was carried out. All measurements of pacing parameters were made by using a pacing system analyzer (Model 2290, Medtronic, Minneapolis, MN).
Study protocol Coronary sinus (CS) cannulation occurred with a commercially available 9-F delivery system. On detection of a PNS o10 V at 0.5 ms during LV lead placement, the LV lead was replaced with a commercially available 5-F decapolar electrophysiology (EP) catheter (Torqr CS Model 041590CS, Medtronic) in which interelectrode spacing is 2–5–2 mm, covering a span of about 40 mm in length. This occurred in 2 steps: location of the site with the lowest phrenic stimulation threshold and EP catheter placement. Location of the phrenic stimulation site The coronary vein was selectively cannulated by means of a commercially available 7-F vein subselection catheter (Attain Select II, Medtronic) to minimize the risk of damaging the coronary vein while advancing the EP catheter. The 5-F EP lead was placed in the vein to wedge distally as
far as possible, and the Attain Select II subselection catheter was pulled back to uncover the 10 electrodes (Figure 1). Then, the PNS threshold was measured in the unipolar configuration to identify the site with the lowest PNS threshold.
EP catheter placement The EP catheter tip was placed at the lowest PNS threshold site in the target coronary vein. This was a trade-off with catheter stability owing to the vein anatomy and to the shifting during the cardiac cycle. Once the EP catheter placement was set, the study measurements were carried out: PNS and LVP thresholds were measured in the unipolar configuration by using each EP lead electrode as the LV cathode and the RV lead coil as the LV anode. Threshold measurements were performed by pacing 10 beats/min above the intrinsic heart rate and stepping down from 10 V at 0.5 ms pulse width in 0.1 V steps until loss of either PNS or LVP capture. When one of these was observed, stimulation was resumed at the previous step, showing PNS or LVP capture, and a stepping down was performed in 0.1 V steps. PNS threshold was defined as the lowest voltage eliciting detectable PNS, though intermittent, during a 20-second observation period.8 LVP threshold was defined as the lowest voltage showing 100% capture. The electrocardiogram was monitored during the LVP threshold measurement to rule out anodal capture. The cathode on the EP catheter with the lowest PNS threshold in the unipolar configuration was identified as the targeted phrenic nerve electrode (TPNE), which was considered to be closest to the phrenic nerve. Once unipolar measurements were carried out, bipolar PNS and LVP threshold measurements were made by using the TPNE as the LVP cathode. Each of the other EP electrodes served as the LVP anode in turn. In this way, we assessed the effect of bipolar electrode spacing on both PNS and LVP thresholds.
Biffi et al Table 1
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
41
Comparison of baseline clinical parameters between groups
Parameter Sex: Man, n (%) Age (y) Ischemic etiology, n (%) Valvular disease, n (%) NYHA class, n (%) II III IV QRS duration (ms) LV ejection fraction (%) LVEDV (mL) LVESV (mL) Atrial fibrillation/atrial flutter, n (%) Persistent/permanent, n (%) LV lead in lateral or posterolateral vein, n (%)
Total patients (N ¼ 40)
Patient with PNS (N ¼ 23)
Patient without PNS (N ¼ 17)
P
29 (73) 16 (40) 18 (45)
15 (65) 66 ⫾ 12 8 (35) 8 (35)
14 (82) 66 ⫾ 9 8 (50) 10 (62)
.297 1.000 .433 .131
18 (45) 21 (53) 1 (3) 158 ⫾ 30 30 ⫾ 10 191 ⫾ 54 137 ⫾ 45 16 (40) 7 (18) 25 (63)
9 (39) 14 (61) 0 (0) 165 ⫾ 27 31 ⫾ 8 176 ⫾ 53 124 ⫾ 43 10 (43) 4 (17) 18 (78)
9 (56) 7 (44) 1 (6) 148 ⫾ 32 29 ⫾ 6 218 ⫾ 49 158 ⫾ 40 6 (38) 3 (19) 7 (44)
.385 .218 .425 .081 .400 .032 .036 .601 1.000 .05
LV ¼ left ventricular; LVEDV ¼ left ventricular end diastolic volume; LVESV ¼ left ventricular end systolic volume; NYHA ¼ New York Heart Association; PNS ¼ phrenic nerve stimulation.
PNS and LVP thresholds were also measured by delivering bipolar pacing from the 3 consecutive 2-mm-spaced dipoles closest to the TPNE in both available polarities. Owing to this specific experimental setting, there was a mismatch between the coronary vein size (which had to accommodate a 7-F subselection catheter) and the 5-F EP catheter electrode, which might result in higher LVP threshold. However, for study purposes, we were more interested in the effect of bipolar electrode spacing on the PNS and LVP threshold values than in the LVP threshold absolute value.
Statistical analysis Continuous variables were shown as mean ⫾ standard deviation, while categorical ones were shown as absolute and relative frequency. Comparisons of baseline parameters between patients with and without PNS were applied by using the Student t test or nonparametric test for continuous distributions and the w2 or Fisher exact test for categorical data. The normality of distributions was assessed by using the ShapiroWilk test. To assess the effect of the distance from the TPNE as well as the bipolar electrode spacing on the PNS and LVP threshold values, the repeated measures ANOVA test was used. Because of the skewness of data, the trends of PNS and LVP thresholds were analyzed by applying the log-transformed distributions. If compound symmetry in the variancecovariance assumption was violated, Box’s conservative correction was used to assess the significance level. A 2-sided P value of r.05 was considered statistically significant. All analysis was performed by using the Stat 12.1 (StataCorp, College Station, TX) statistical package.
Hence, only 23 of 40 (57%) patients completed the study protocol. The clinical characteristics of the 40 patients and of the 23 who completed the study are listed in Table 1. In this experimental setting, the phrenic nerve was located at the distal end of the decapolar catheter in all but one patient. The EP catheter tip was placed at the site with the lowest PNS threshold in 15 of 23 (65.2%) patients; electrode 2 was the TPNE in 4 of 23 (17.4%) patients; electrode 4 was the TPNE in 2 of 23 (8.7%) patients; electrodes 5 and 9 were the TPNE in 1 (8.7%) patient each. This occurred because of catheter stability along the vein course in 6 patients and because of the vein length in 1 patient.
Effect of the cathode distance from the phrenic nerve in the unipolar configuration The PNS threshold at the TPNE was, on average, lower than the LVP threshold (Figure 2). We observed that the PNS threshold increased with distance from the phrenic nerve, being minimal at the cathode closest to the nerve (P o .0001). The LVP threshold had little change (P ¼ .0584) and appeared to be increasing at more proximal pacing sites owing to larger vein diameter (poor electrode contact). In 15 patients, we compared the PNS threshold and the LVP threshold, as measured by means of the implanted LV lead with the tip at the same TPNE site in the coronary vein, with those from the EP catheter: the latter underestimated the unipolar LVP threshold, possibly owing to the different mechanical properties and contact force (Table 2). Indeed, the effect of the distance from the TPNE on both unipolar PNS and LVP thresholds was similar for the implanted LV lead compared to the EP catheter (Table 2).
Results Forty patients signed the informed consent in the study period: 14 did not have PNS when paced at outputs lower than 10 V at 0.5 ms, and 3 could not complete the study protocol because no PNS occurred in bipolar configurations.
Effect of bipolar electrode spacing on bipolar PNS and LVP thresholds We observed that shortening the bipolar electrode spacing when the LVP cathode is at the TPNE increases the PNS
42
Heart Rhythm, Vol 10, No 1, January 2013
Figure 3 Effect of bipolar interelectrode spacing on bipolar phrenic nerve stimulation (PNS) threshold (A) and on bipolar left ventricular pacing (LVP) threshold (B). Figure 2 Effect of the cathode distance from the phrenic nerve on both phrenic nerve stimulation (PNS) threshold (A) and unipolar left ventricular pacing (LVP) threshold (B). TPNE ¼ targeted phrenic nerve electrode.
threshold (P ¼ .019; Figure 3A) while no difference was found for the LVP threshold (P ¼ .4278; Figure 3B). Excluding the 40 mm spacing, the PNS threshold is, on average, 1 V higher at a 2 mm spacing compared to all other bipolar configurations. Table 2
The comparison of the implanted LV lead with the EP catheter in 15 patients showed that the latter underestimated the advantage of interelectrode spacing in the 20 mm length (Table 2).
Effect of the distance of a short-dipole location from the TPNE The PNS threshold markedly increased when pacing was delivered by a short-spaced dipole (2 mm) placed 5 mm
Comparison of EP catheter-based and LV lead-based measurements of PNS and LVP thresholds at the same pacing sites P
Pacing configuration
EP catheter
LV lead
EP catheter vs LV lead
PNS threshold (V at 0.5 ms) TPNE to RV coil Electrode 20 mm from TPNE to RV coil Bipolar 20 mm spaced
1.10 ⫾ 0.60 7.63 ⫾ 3.06 2.14 ⫾ 1.44
2.80 ⫾ 3.60 6.40 ⫾ 3.20 4.70 ⫾ 3.60
LVP hreshold (V at 0.5 ms) TPNE to RV coil Electrode 20 mm from TPNE to RV coil Bipolar 20 mm spaced
2.22 ⫾ 2.22 2.40 ⫾ 2.29 2.42 ⫾ 2.00
1.40 ⫾ 1.30 1.00 ⫾ 0.50 2.32 ⫾ 2.00
EP catheter
LV lead
.032 .278 .005
o.0001
.007
—
—
.11 .027 .089
.788
.276
—
—
Values are given as mean ⫾ standard deviation. EP ¼ electrophysiology; LV ¼ left ventricular; LVP ¼ left ventricular pacing; PNS ¼ phrenic nerve stimulation; RV ¼ right ventricular; TPNE ¼ targeted phrenic nerve electrode.
Biffi et al
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
Figure 4 Effect of the cathode distance from the phrenic nerve of a 2-mm-spaced dipole on bipolar phrenic nerve stimulation (PNS) threshold (square) and left ventricular pacing (LVP) threshold (diamond). TPNE ¼ targeted phrenic nerve electrode.
away from the TPNE (Figure 4). The average PNS threshold changed a little at progressively increased distances of the short-spaced dipole from the TPNE (P ¼ .026). On the other hand, no significant changes in the LVP threshold were detected (P ¼ .119). Moreover, the percentage of patients who avoided PNS (PNS threshold Z10 V at 0.5 ms) increased when the 2-mm-spaced dipole was located away from the TPNE in comparison with unipolar pacing at the same distance (Figure 5). The percentage of patients without PNS further increased even in moving the cathode 2 mm further from the TPNE by reversing the polarity of the dipole (Figure 5).
Discussion The main finding of our study is that both the location of the LVP cathode with respect to the phrenic nerve and the bipolar electrode spacing of the pacing dipole can affect the PNS threshold. This is based on our data that the chance to elicit PNS decreases at increasing distance from the phrenic nerve, or when shortening the bipolar electrode spacing.
Figure 5 Phrenic nerve stimulation (PNS) avoidance: probability of freedom from PNS at a pacing output of Z10 V at 0.5 ms in unipolar (black), 2-mm-spaced bipolar (gray), and 2-mm-spaced bipolar with reverse polarity (striped) configurations. TPNE ¼ targeted phrenic nerve electrode.
43
When a short-spaced dipole (2 mm) is placed between 5 and 25 mm far from the site with the lowest PNS threshold, avoidance of PNS is achieved in 80%–100% of the cases. This has several important implications for CRT in clinical practice. In fact, PNS need to be managed at the same pacing site that is deemed optimal for CRT, where the incidence of PNS is maximal.8–13 When PNS is detected at implantation, several approaches are used: to move the lead to a more proximal site, to program the LVP cathode other than the lead tip in bipolar/multipolar leads featuring cathode/anode programmability,8–13,16,17 or to lower the LVP output so as to avoid PNS when other options have failed.8–14 LV lead repositioning at another site is the last resort, and it is possible only when coronary veins other than the target one are suitable for LV lead placement. Each of these approaches has its own drawbacks: LV lead placement at a proximal site poses an increased risk of suboptimal LVP threshold and of LV lead displacement, requiring additional surgery 8,10 that might lead to infective complications,18 whereas abandoning the target site may cause failure to deliver effective CRT and accompanying clinical improvements.4–7 Our observations about these 2 strategies for PNS avoidance, namely, placement of the LVP cathode remotely from the TPNE or shortening the bipolar electrode spacing while keeping the LVP cathode close to the TPNE, can potentially affect the LV lead design and manufacturing processes.
Effect of the cathode distance from the phrenic nerve in the unipolar configuration In the unipolar configuration, the PNS threshold at the TPNE was, on average, lower than the LVP threshold (Figure 2). This may be partly related to the proximity of the TPNE to the phrenic nerve, according to the protocol, and to poor contact owing to vein size. In addition, the PNS threshold increased with distance from the phrenic nerve (Figure 2) while the LVP threshold had little changes. An increase in the LVP threshold at distances 430 mm was presumably due to the mismatch between the size of the proximal portion of the vein and the 5-F EP catheter. When close to the atrioventricular groove, coronary veins may be embedded in fat tissue, which also causes a higher than conventional LVP threshold. LV stimulation at a site far from the TPNE is the strategy most commonly used in clinical practice to avoid PNS. The feature of LVP cathode reprogramming accomplishes this task at no increased risk of lead displacement,11,12,16 as the lead is not moved to a more proximal site.8–10 This strategy has not been proven to ensure PNS avoidance in 100% of the patients with bipolar LV leads.11,12,14,15 Although clinically effective to provide relief from PNS symptoms, the new CRT defibrillator equipped with a quadripolar LV lead cannot comprehensively address the PNS issue: during implantation in highly trained centers, 5 of 75 (6.6%) patients had the lead placed at a vein different from the target one, because PNS could not be managed (16).
44 Moreover, the incidence of PNS at 7.5 V when programming the LVP cathode at 30 or 47 mm from the lead tip was 14% and 23%, respectively, with an average PNS threshold around 5 ⫾ 2 V.16 The LVP threshold in those settings was, respectively, 2.5 and 3.5 V, on average; this means that a difference of 3 V between the PNS threshold and the LVP threshold, which has been proven effective as an implantation criterion to warrant freedom from PNS-related complications in more than 700 patients,8,9,12,13,19,20 was not warranted in all patients. Beyond representing a potential limitation to managing PNS at follow-up, these observations highlight that the management of PNS by LVP cathode reprogramming may be a trade-off with a high LVP threshold in about 25% of the patients, as recently reported by Klein et al.17 Hence, owing to the variability of both coronary vein and phrenic nerve anatomy, a strategy based on a multielectrode LV lead at conventional dipole electrode spacing (minimum dipole electrode spacing is 10 mm) does not seem to provide a comprehensive approach to PNS.19,20
Effect of electrode spacing on the PNS threshold and implications for a PNS management strategy In multielectrode LV leads, the length of electrode spacing is also important when addressing PNS. Stimulation at the TPNE with a 2-mm-spaced dipole increased the PNS threshold compared to other bipolar spacings (Figure 3). On the contrary, the LVP threshold was not significantly affected by the dipole length. Owing to poor contact with the myocardium and to the EP electrode surface area in our settings, measured LVP thresholds do not match those measured in clinical practice with standard LVP leads. However, for the study purposes, we were more interested in the PNS to LVP threshold relationship than in the LV threshold absolute value. Most importantly, we were able to locate the distance of a 2-mm-spaced dipole from the TPNE that might warrant a very high chance of PNS avoidance (Figures 4 and 5). Indeed, 100% of the patients could avoid PNS at 10 V when stimulation occurred between 15 and 25 mm from the TPNE (Figure 5). The LVP threshold was unchanged while PNS increased; the PNS to LVP safety margin increased to 45 V when a short dipole was positioned 5 mm or further from the phrenic nerve (Figure 4), fully acceptable for managing PNS long-term in clinical practice. As reported above, a 3 V safety margin has been reported as the least efficient to manage the PNS issue.8,9,12,13 The role of a short-spaced dipole is strongly supported by experimental evidence in a canine model, as recently reported.21,22 Wecke et al21 demonstrated that a very short-spaced dipole (r1 mm in length), obtained by pacing between adjacent segments of networked electrodes, can achieve PNS avoidance in 100% of worst-case scenarios (phrenic nerve dissected and positioned above the LV lead). Our observations support the concept that multipolar LV leads should have a short-spaced dipole (1–2 mm in length) to ensure PNS management at follow-up. The location of a
Heart Rhythm, Vol 10, No 1, January 2013 short-spaced dipole respective to the lead tip might be different, depending on vein length and size, as PNS occurs most commonly at mid-to-distal posterior and lateral sites.8,9 Hence, because the optimal LV pacing site to ensure maximum mechanical improvement may have substantial individual variability,4,23 the placement of a short-spaced dipole should vary as well, from close to the lead tip to 20–25 mm proximal from the lead tip. A new family of leads would allow the implanting physician to keep the LV lead in the target site for CRT, despite being very close to the phrenic nerve. This would improve clinical outcome of CRT response5–7,24 and reduce the risk of significant morbidity and mortality.1,2,4–7,18
Conclusions and practical implications We believe that our study provides the rationale for the development of multielectrode LV leads with 1–2 mm bipolar electrode spacing. This lead design would maximize PNS management, both at implantation and at follow-up, while allowing individual targeting of the LV pacing site.
Study limitations Contrary to standard clinical practice, we used commercially available EP catheters. Thus, we cannot ensure that the same observations about short-spaced dipole stimulation can be transferred to the clinical practice where different leads are employed. The EP catheter had indeed an increased LV pacing threshold compared to conventional LV pacing leads in our patients, possibly owing to the different mechanical properties and to unpredictable electrode contact with the myocardium, as previously reported in an animal study.22 It is therefore possible that the extent of the benefit of stimulation by a short-spaced dipole is underestimated. Moreover, this is an acute study that cannot provide evidence of consistency of our findings in the long-term. Chronic studies will be needed to confirm the possibility of managing PNS at the selected LV pacing site once shortspaced dipole LV leads are manufactured.
Appendix Participating investigators and their affiliations are listed here in alphabetical order: Battistini P, Biffi M, Boriani G, Diemberger I, Martignani C, Ziacchi M, Institute of Cardiology, University of Bologna, Bologna, Italy; Baracca E, Pastore G, Zanon F, Zerbo F, Ospedale Santa Maria della Misericordia, Rovigo, Italy; Bertaglia E, Zoppo F, Ospedale Civile, Mirano, Italy; Padeletti L, Perrotta L, Pieragnoli P, Ricciardi G, University of Florence, Florence, Italy; De Santo T, Varbaro A, Medtronic Italia SpA, Italy; Yang Z, Medtronic Inc, Minneapolis, Minnesota.
Acknowledgments The authors thank Valeria Ricci and Giovanna Zucchi from Medtronic, Italy, for their helpful participation in data collection.
Biffi et al
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
References 1. Bristow MR, Saxon LA, Boehmer J, et al, Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140–2150. 2. Cleland JG, Daubert JC, Erdmann E, et al, for the Cardiac ResynchronizationHeart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: 1539–1549. 3. Butter C, Auricchio A, Stellbrink C, et al, on behalf of the Pacing Therapy for Chronic Heart Failure II (PATH-CHF-II) Study Group. Effect of resynchronization therapy stimulation site on the systolic function of heart failure patients. Circulation 2001;104:3026–3029. 4. Dekker AL, Phelps B, Dijkman B, et al, Epicardial left ventricular lead placement for cardiac resynchronization therapy: optimal pace site selection with pressurevolume loops. J Thorac Cardiovasc Surg 2004;127:1641–1647. 5. Ypenburg C, van Bommel RJ, Delgado V, et al, Optimal left ventricular lead position predicts reverse remodeling and survival after cardiac resynchronization Therapy. J Am Coll Cardiol 2008;52:1402–1409. 6. Ypenburg C, van Bommel RJ, Borleffs CJW, et al, Long-term prognosis after cardiac resynchronization therapy is related to the extent of left ventricular reverse remodeling at midterm follow-up. J Am Coll Cardiol 2009;53:483–490. 7. Khan FZ, Virdee MS, Palmer CR, et al, Targeted left ventricular lead placement to guide cardiac resynchronization therapy. The TARGET study: a randomized, controlled trial. J Am Coll Cardiol 2012;59:1509–1518. 8. Biffi M, Moschini C, Bertini M, et al, Phrenic stimulation: a challenge for cardiac resynchronization therapy. Circ Arrhythm Electrophysiol 2009;2:402–410. 9. Seifert M, Schau T, Moeller V, Neuss M, Meyhoefer J, Butter C. Influence of pacing configurations, body mass index, and position of coronary sinus lead on frequency of phrenic nerve stimulation and pacing thresholds under cardiac resynchronization therapy. Europace 2010;12:961–967. 10. Gurevitz O, Nof E, Carasso S, et al, Programmable multiple pacing configurations help to overcome high left ventricular pacing thresholds and avoid phrenic nerve stimulation. Pacing Clin Electrophysiol 2005;28:1255–1259. 11. Champagne J, Healey JS, Krahn AD, et al, ELECTION Investigators. The effect of electronic repositioning on left ventricular pacing and phrenic nerve stimulation. Europace 2011;13:409–415. 12. Biffi M, Bertini M, Ziacchi M, et al, Management of phrenic stimulation in CRT patients over the long term: still an unmet need? Pacing Clin Electrophysiol 2011;34:1201–1208.
45
13. Jastrzebski M, Bacior B, Wojciechowska W, Czarnecka W. Left ventricular lead implantation at a phrenic stimulation site is safe and effective. Europace 2011;13: 520–525. 14. Biffi M, Bertini M, Saporito D, et al, Automatic management of left ventricular stimulation: hints for technologic improvement. Pacing Clin Electrophysiol 2009;32:346–353. 15. Biffi M, Boriani G. Phrenic stimulation management in CRT patients: are we there yet?. Curr Opin Cardiol 2011;26:12–16. 16. Sperzel J, D¨anschel W, Gutleben KJ, et al, First prospective, multi-centre clinical experience with a novel left ventricular quadripolar lead. Europace 2012;14: 365–372. 17. Klein N, Klein M, Weglage H, et al, Clinical efficacy of left ventricular pacing vector programmability in cardiac resynchronization therapy defibrillator patients for management of phrenic nerve stimulation and/or elevated left ventricular pacing thresholds: insights from the Efface Phrenic Stim study. Europace 2012;14:826–832. 18. Poole JE, Gleva MJ, Mela T, et al, for the REPLACE Registry Investigators. Complication rates associated with pacemaker or implantable cardioverterdefibrillator generator replacements and upgrade procedures. Circulation 2010;122:1553–1561. 19. Kirubakaran S, Rinaldi CA. Phrenic nerve stimulation with the quadripolar left ventricular lead not overcome by “electronic repositioning.” Europace 2012;14: 608–609. 20. Parahuleva M, Chasan R, Soydan N, et al, Quadripolar left ventricular lead in a patient with CRT-D does not overcome phrenic nerve stimulation. Clin Med Insights Cardiol 2011;5:45–47. 21. Wecke L, van Hunnik A, Thompson T, et al, Networked multi-electrode left ventricular pacing lead for avoidance of phrenic nerve stimulation in a canine model. Heart Rhythm 2012;9:789–795. 22. Biffi M, Foerster L, Eastman W, et al, Effect of bipolar electrode spacing on phrenic nerve stimulation and left ventricular pacing thresholds: an acute canine study. Circ Arrhythm Electrophysiol 2012;5:815–820. 23. Derval N, Steendijk P, Gula LJ, et al, Optimizing hemodynamics in heart failure patients by systematic screening of left ventricular pacing sites: the lateral left ventricular wall and the coronary sinus are rarely the best sites. J Am Coll Cardiol 2010;55:566–575. 24. Macı´as A, Gavira J-J, Castan˜o S, Alegrı´a E, Garcı´a-Bolao I. Left ventricular pacing site in cardiac resynchronization therapy: clinical follow-up and predictors of failed lateral implant. Eur J Heart Fail 2008;10:421–427.
Short-spaced dipole for managing phrenic nerve stimulation in patients with CRT: The “phrenic nerve mapping and stimulation EP” catheter study Mauro Biffi, MD,* Francesco Zanon, MD,† Emanuele Bertaglia, MD,‡ Luigi Padeletti, MD, PhD,y Annamaria Varbaro, MS,k Tiziana De Santo, MSc,k Giuseppe Boriani, MD, PhD,* Zhongping Yang, PhDz on behalf of the PING-EP Study Group From the *Institute of Cardiology, University of Bologna, Bologna, Italy, yOspedale Santa Maria della Misericordia, Rovigo, Italy, zOspedale Civile, Mirano, Italy, yUniversity of Florence, Florence, Italy, kMedtronic Italia SpA, Milano, Italy and z Medtronic Inc, Minneapolis, Minnesota. BACKGROUND Phrenic nerve stimulation (PNS), occurring in 33%–37% of the patients with cardiac resynchronization therapy (CRT), is a limiting factor when implanting left ventricular (LV) leads from coronary veins. OBJECTIVE To test the hypothesis that PNS occurence is related to bipolar electrode spacing. METHODS During standard CRT defibrillator implant procedures, a 5-F diagnostic electrophysiology catheter with 10 electrodes, spaced 2–5–2 mm, was positioned in a cardiac vein suitable for permanent LV lead placement. Pacing in the unipolar configuration identified the site with the lowest PNS threshold. PNS and left ventricular pacing (LVP) thresholds were then measured in different configurations at 0.5 ms: unipolar, each LV electrode served as the cathode in turn; and bipolar with different electrode spacing, cathode being the electrode with the lowest unipolar PNS threshold. RESULTS From February to September 2010, 40 patients undergoing CRT implantation were enrolled in 4 centers in Italy. It was possible to identify PNS and perform a complete set of
measurements in 23 patients. A bipolar electrode spacing of 2 mm resulted in higher PNS thresholds in bipolar configurations han did a bipolar electrode spacing of Z5 mm. However, no significant increase in the LVP threshold was observed (P ¼ ns). CONCLUSIONS This experience suggests that LVP with a bipolar electrode spacing of 2 mm significantly increases the PNS threshold without affecting the LVP threshold, thereby increasing the possibility of delivering CRT when the LV lead is placed in proximity to the phrenic nerve. KEYWORDS Cardiac resynchronization therapy; Phrenic nerve stimulation; Bipolar electrode spacing; Threshold ABBREVIATIONS CRT ¼ cardiac resynchronization therapy; EP ¼ electrophysiology; LV ¼ left ventricular; LVP ¼ left ventricular pacing; PNS ¼ phrenic nerve stimulation; RV ¼ right ventricular; TPNE ¼ targeted phrenic nerve electrode (Heart Rhythm 2013;10:39–45) I 2013 The Heart Rhythm Society. All rights reserved.
3 3
Introduction Cardiac resynchronization therapy (CRT) is an established treatment of heart failure caused by left ventricular (LV) systolic dysfunction with evidence of electrical/mechanical dyssynchrony.1,2 CRT aims at improving the LV mechanics by stimulating the mostly delayed LV sites.3–5 This approach has been shown to result in clinical benefits.5–7 Phrenic nerve stimulation (PNS) is a challenge to delivering effective CRT. It has been observed in 33%–37% of the patients8–10 and is more common at the same LV sites where the chance of a Dr Biffi received a modest honorarium as speaker’s bureau and education activity. Dr Yang, Ms Varbaro, and Ms Santo are employees of Medtronic. Address reprint requests and correspondence: Dr Mauro Biffi, MD, Institute of Cardiology, University of Bologna, Via Massarenti 9, 40138 Bologna, Italy. E-mail address: [email protected].
favorable response to CRT is maximal.5,8 Although PNS is actively addressed at implantation,8 the long-term management of PNS may present a challenge in patients with CRT.11–13 Indeed, about 15% of the patients need to be reevaluated because of PNS occurrence at follow-up11–13 and about 6.6% eventually report long-term PNS symptoms despite repeated attempts at PNS avoidance.12 Moreover, the absence of PNS at implantation does not ensure freedom from PNS at follow-up.8–10 Although built-in features, for example, automated LV output adjustment and capability to separately program LV lead cathode/anode can help manage PNS,8,10,11,14 a definitive solution to this challenge is not available.15 We sought to investigate the physiologic principles of PNS in an acute study of CRT candidates in order to develop a comprehensive strategy aimed at avoiding PNS symptoms in clinical practice.
1547-5271/$-see front matter B 2013 The Heart Rhythm Society. All rights reserved.
http://dx.doi.org/10.1016/j.hrthm.2012.08.045
40
Heart Rhythm, Vol 10, No 1, January 2013
Figure 1 Fluoro image of coronary vein angiogram and of electrophysiology catheter placed into the coronary vein. Lead tip is the targeted phrenic nerve electrode (TPNE).
Methods We investigated the relationship between the distance of the left ventricular pacing (LVP) cathode from the phrenic nerve and the PNS threshold in the unipolar configuration (LV cathode to RV coil) as well as the relationship between bipolar electrode spacing and the PNS threshold in the true bipolar configuration (LV cathode to LV anode) when the LV cathode was placed at the site with the lowest PNS threshold. This study was carried out at 4 Italian hospitals with more than 8 years of experience in CRT implantation. The study protocol was approved by the hospital ethical committee of each center. During February to September 2010, patients scheduled for a CRT implantation were asked to participate in this clinical study by signing an informed consent: in the event PNS was detected during LV lead implantation, a detailed study of PNS was carried out. All measurements of pacing parameters were made by using a pacing system analyzer (Model 2290, Medtronic, Minneapolis, MN).
Study protocol Coronary sinus (CS) cannulation occurred with a commercially available 9-F delivery system. On detection of a PNS o10 V at 0.5 ms during LV lead placement, the LV lead was replaced with a commercially available 5-F decapolar electrophysiology (EP) catheter (Torqr CS Model 041590CS, Medtronic) in which interelectrode spacing is 2–5–2 mm, covering a span of about 40 mm in length. This occurred in 2 steps: location of the site with the lowest phrenic stimulation threshold and EP catheter placement. Location of the phrenic stimulation site The coronary vein was selectively cannulated by means of a commercially available 7-F vein subselection catheter (Attain Select II, Medtronic) to minimize the risk of damaging the coronary vein while advancing the EP catheter. The 5-F EP lead was placed in the vein to wedge distally as
far as possible, and the Attain Select II subselection catheter was pulled back to uncover the 10 electrodes (Figure 1). Then, the PNS threshold was measured in the unipolar configuration to identify the site with the lowest PNS threshold.
EP catheter placement The EP catheter tip was placed at the lowest PNS threshold site in the target coronary vein. This was a trade-off with catheter stability owing to the vein anatomy and to the shifting during the cardiac cycle. Once the EP catheter placement was set, the study measurements were carried out: PNS and LVP thresholds were measured in the unipolar configuration by using each EP lead electrode as the LV cathode and the RV lead coil as the LV anode. Threshold measurements were performed by pacing 10 beats/min above the intrinsic heart rate and stepping down from 10 V at 0.5 ms pulse width in 0.1 V steps until loss of either PNS or LVP capture. When one of these was observed, stimulation was resumed at the previous step, showing PNS or LVP capture, and a stepping down was performed in 0.1 V steps. PNS threshold was defined as the lowest voltage eliciting detectable PNS, though intermittent, during a 20-second observation period.8 LVP threshold was defined as the lowest voltage showing 100% capture. The electrocardiogram was monitored during the LVP threshold measurement to rule out anodal capture. The cathode on the EP catheter with the lowest PNS threshold in the unipolar configuration was identified as the targeted phrenic nerve electrode (TPNE), which was considered to be closest to the phrenic nerve. Once unipolar measurements were carried out, bipolar PNS and LVP threshold measurements were made by using the TPNE as the LVP cathode. Each of the other EP electrodes served as the LVP anode in turn. In this way, we assessed the effect of bipolar electrode spacing on both PNS and LVP thresholds.
Biffi et al Table 1
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
41
Comparison of baseline clinical parameters between groups
Parameter Sex: Man, n (%) Age (y) Ischemic etiology, n (%) Valvular disease, n (%) NYHA class, n (%) II III IV QRS duration (ms) LV ejection fraction (%) LVEDV (mL) LVESV (mL) Atrial fibrillation/atrial flutter, n (%) Persistent/permanent, n (%) LV lead in lateral or posterolateral vein, n (%)
Total patients (N ¼ 40)
Patient with PNS (N ¼ 23)
Patient without PNS (N ¼ 17)
P
29 (73) 16 (40) 18 (45)
15 (65) 66 ⫾ 12 8 (35) 8 (35)
14 (82) 66 ⫾ 9 8 (50) 10 (62)
.297 1.000 .433 .131
18 (45) 21 (53) 1 (3) 158 ⫾ 30 30 ⫾ 10 191 ⫾ 54 137 ⫾ 45 16 (40) 7 (18) 25 (63)
9 (39) 14 (61) 0 (0) 165 ⫾ 27 31 ⫾ 8 176 ⫾ 53 124 ⫾ 43 10 (43) 4 (17) 18 (78)
9 (56) 7 (44) 1 (6) 148 ⫾ 32 29 ⫾ 6 218 ⫾ 49 158 ⫾ 40 6 (38) 3 (19) 7 (44)
.385 .218 .425 .081 .400 .032 .036 .601 1.000 .05
LV ¼ left ventricular; LVEDV ¼ left ventricular end diastolic volume; LVESV ¼ left ventricular end systolic volume; NYHA ¼ New York Heart Association; PNS ¼ phrenic nerve stimulation.
PNS and LVP thresholds were also measured by delivering bipolar pacing from the 3 consecutive 2-mm-spaced dipoles closest to the TPNE in both available polarities. Owing to this specific experimental setting, there was a mismatch between the coronary vein size (which had to accommodate a 7-F subselection catheter) and the 5-F EP catheter electrode, which might result in higher LVP threshold. However, for study purposes, we were more interested in the effect of bipolar electrode spacing on the PNS and LVP threshold values than in the LVP threshold absolute value.
Statistical analysis Continuous variables were shown as mean ⫾ standard deviation, while categorical ones were shown as absolute and relative frequency. Comparisons of baseline parameters between patients with and without PNS were applied by using the Student t test or nonparametric test for continuous distributions and the w2 or Fisher exact test for categorical data. The normality of distributions was assessed by using the ShapiroWilk test. To assess the effect of the distance from the TPNE as well as the bipolar electrode spacing on the PNS and LVP threshold values, the repeated measures ANOVA test was used. Because of the skewness of data, the trends of PNS and LVP thresholds were analyzed by applying the log-transformed distributions. If compound symmetry in the variancecovariance assumption was violated, Box’s conservative correction was used to assess the significance level. A 2-sided P value of r.05 was considered statistically significant. All analysis was performed by using the Stat 12.1 (StataCorp, College Station, TX) statistical package.
Hence, only 23 of 40 (57%) patients completed the study protocol. The clinical characteristics of the 40 patients and of the 23 who completed the study are listed in Table 1. In this experimental setting, the phrenic nerve was located at the distal end of the decapolar catheter in all but one patient. The EP catheter tip was placed at the site with the lowest PNS threshold in 15 of 23 (65.2%) patients; electrode 2 was the TPNE in 4 of 23 (17.4%) patients; electrode 4 was the TPNE in 2 of 23 (8.7%) patients; electrodes 5 and 9 were the TPNE in 1 (8.7%) patient each. This occurred because of catheter stability along the vein course in 6 patients and because of the vein length in 1 patient.
Effect of the cathode distance from the phrenic nerve in the unipolar configuration The PNS threshold at the TPNE was, on average, lower than the LVP threshold (Figure 2). We observed that the PNS threshold increased with distance from the phrenic nerve, being minimal at the cathode closest to the nerve (P o .0001). The LVP threshold had little change (P ¼ .0584) and appeared to be increasing at more proximal pacing sites owing to larger vein diameter (poor electrode contact). In 15 patients, we compared the PNS threshold and the LVP threshold, as measured by means of the implanted LV lead with the tip at the same TPNE site in the coronary vein, with those from the EP catheter: the latter underestimated the unipolar LVP threshold, possibly owing to the different mechanical properties and contact force (Table 2). Indeed, the effect of the distance from the TPNE on both unipolar PNS and LVP thresholds was similar for the implanted LV lead compared to the EP catheter (Table 2).
Results Forty patients signed the informed consent in the study period: 14 did not have PNS when paced at outputs lower than 10 V at 0.5 ms, and 3 could not complete the study protocol because no PNS occurred in bipolar configurations.
Effect of bipolar electrode spacing on bipolar PNS and LVP thresholds We observed that shortening the bipolar electrode spacing when the LVP cathode is at the TPNE increases the PNS
42
Heart Rhythm, Vol 10, No 1, January 2013
Figure 3 Effect of bipolar interelectrode spacing on bipolar phrenic nerve stimulation (PNS) threshold (A) and on bipolar left ventricular pacing (LVP) threshold (B). Figure 2 Effect of the cathode distance from the phrenic nerve on both phrenic nerve stimulation (PNS) threshold (A) and unipolar left ventricular pacing (LVP) threshold (B). TPNE ¼ targeted phrenic nerve electrode.
threshold (P ¼ .019; Figure 3A) while no difference was found for the LVP threshold (P ¼ .4278; Figure 3B). Excluding the 40 mm spacing, the PNS threshold is, on average, 1 V higher at a 2 mm spacing compared to all other bipolar configurations. Table 2
The comparison of the implanted LV lead with the EP catheter in 15 patients showed that the latter underestimated the advantage of interelectrode spacing in the 20 mm length (Table 2).
Effect of the distance of a short-dipole location from the TPNE The PNS threshold markedly increased when pacing was delivered by a short-spaced dipole (2 mm) placed 5 mm
Comparison of EP catheter-based and LV lead-based measurements of PNS and LVP thresholds at the same pacing sites P
Pacing configuration
EP catheter
LV lead
EP catheter vs LV lead
PNS threshold (V at 0.5 ms) TPNE to RV coil Electrode 20 mm from TPNE to RV coil Bipolar 20 mm spaced
1.10 ⫾ 0.60 7.63 ⫾ 3.06 2.14 ⫾ 1.44
2.80 ⫾ 3.60 6.40 ⫾ 3.20 4.70 ⫾ 3.60
LVP hreshold (V at 0.5 ms) TPNE to RV coil Electrode 20 mm from TPNE to RV coil Bipolar 20 mm spaced
2.22 ⫾ 2.22 2.40 ⫾ 2.29 2.42 ⫾ 2.00
1.40 ⫾ 1.30 1.00 ⫾ 0.50 2.32 ⫾ 2.00
EP catheter
LV lead
.032 .278 .005
o.0001
.007
—
—
.11 .027 .089
.788
.276
—
—
Values are given as mean ⫾ standard deviation. EP ¼ electrophysiology; LV ¼ left ventricular; LVP ¼ left ventricular pacing; PNS ¼ phrenic nerve stimulation; RV ¼ right ventricular; TPNE ¼ targeted phrenic nerve electrode.
Biffi et al
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
Figure 4 Effect of the cathode distance from the phrenic nerve of a 2-mm-spaced dipole on bipolar phrenic nerve stimulation (PNS) threshold (square) and left ventricular pacing (LVP) threshold (diamond). TPNE ¼ targeted phrenic nerve electrode.
away from the TPNE (Figure 4). The average PNS threshold changed a little at progressively increased distances of the short-spaced dipole from the TPNE (P ¼ .026). On the other hand, no significant changes in the LVP threshold were detected (P ¼ .119). Moreover, the percentage of patients who avoided PNS (PNS threshold Z10 V at 0.5 ms) increased when the 2-mm-spaced dipole was located away from the TPNE in comparison with unipolar pacing at the same distance (Figure 5). The percentage of patients without PNS further increased even in moving the cathode 2 mm further from the TPNE by reversing the polarity of the dipole (Figure 5).
Discussion The main finding of our study is that both the location of the LVP cathode with respect to the phrenic nerve and the bipolar electrode spacing of the pacing dipole can affect the PNS threshold. This is based on our data that the chance to elicit PNS decreases at increasing distance from the phrenic nerve, or when shortening the bipolar electrode spacing.
Figure 5 Phrenic nerve stimulation (PNS) avoidance: probability of freedom from PNS at a pacing output of Z10 V at 0.5 ms in unipolar (black), 2-mm-spaced bipolar (gray), and 2-mm-spaced bipolar with reverse polarity (striped) configurations. TPNE ¼ targeted phrenic nerve electrode.
43
When a short-spaced dipole (2 mm) is placed between 5 and 25 mm far from the site with the lowest PNS threshold, avoidance of PNS is achieved in 80%–100% of the cases. This has several important implications for CRT in clinical practice. In fact, PNS need to be managed at the same pacing site that is deemed optimal for CRT, where the incidence of PNS is maximal.8–13 When PNS is detected at implantation, several approaches are used: to move the lead to a more proximal site, to program the LVP cathode other than the lead tip in bipolar/multipolar leads featuring cathode/anode programmability,8–13,16,17 or to lower the LVP output so as to avoid PNS when other options have failed.8–14 LV lead repositioning at another site is the last resort, and it is possible only when coronary veins other than the target one are suitable for LV lead placement. Each of these approaches has its own drawbacks: LV lead placement at a proximal site poses an increased risk of suboptimal LVP threshold and of LV lead displacement, requiring additional surgery 8,10 that might lead to infective complications,18 whereas abandoning the target site may cause failure to deliver effective CRT and accompanying clinical improvements.4–7 Our observations about these 2 strategies for PNS avoidance, namely, placement of the LVP cathode remotely from the TPNE or shortening the bipolar electrode spacing while keeping the LVP cathode close to the TPNE, can potentially affect the LV lead design and manufacturing processes.
Effect of the cathode distance from the phrenic nerve in the unipolar configuration In the unipolar configuration, the PNS threshold at the TPNE was, on average, lower than the LVP threshold (Figure 2). This may be partly related to the proximity of the TPNE to the phrenic nerve, according to the protocol, and to poor contact owing to vein size. In addition, the PNS threshold increased with distance from the phrenic nerve (Figure 2) while the LVP threshold had little changes. An increase in the LVP threshold at distances 430 mm was presumably due to the mismatch between the size of the proximal portion of the vein and the 5-F EP catheter. When close to the atrioventricular groove, coronary veins may be embedded in fat tissue, which also causes a higher than conventional LVP threshold. LV stimulation at a site far from the TPNE is the strategy most commonly used in clinical practice to avoid PNS. The feature of LVP cathode reprogramming accomplishes this task at no increased risk of lead displacement,11,12,16 as the lead is not moved to a more proximal site.8–10 This strategy has not been proven to ensure PNS avoidance in 100% of the patients with bipolar LV leads.11,12,14,15 Although clinically effective to provide relief from PNS symptoms, the new CRT defibrillator equipped with a quadripolar LV lead cannot comprehensively address the PNS issue: during implantation in highly trained centers, 5 of 75 (6.6%) patients had the lead placed at a vein different from the target one, because PNS could not be managed (16).
44 Moreover, the incidence of PNS at 7.5 V when programming the LVP cathode at 30 or 47 mm from the lead tip was 14% and 23%, respectively, with an average PNS threshold around 5 ⫾ 2 V.16 The LVP threshold in those settings was, respectively, 2.5 and 3.5 V, on average; this means that a difference of 3 V between the PNS threshold and the LVP threshold, which has been proven effective as an implantation criterion to warrant freedom from PNS-related complications in more than 700 patients,8,9,12,13,19,20 was not warranted in all patients. Beyond representing a potential limitation to managing PNS at follow-up, these observations highlight that the management of PNS by LVP cathode reprogramming may be a trade-off with a high LVP threshold in about 25% of the patients, as recently reported by Klein et al.17 Hence, owing to the variability of both coronary vein and phrenic nerve anatomy, a strategy based on a multielectrode LV lead at conventional dipole electrode spacing (minimum dipole electrode spacing is 10 mm) does not seem to provide a comprehensive approach to PNS.19,20
Effect of electrode spacing on the PNS threshold and implications for a PNS management strategy In multielectrode LV leads, the length of electrode spacing is also important when addressing PNS. Stimulation at the TPNE with a 2-mm-spaced dipole increased the PNS threshold compared to other bipolar spacings (Figure 3). On the contrary, the LVP threshold was not significantly affected by the dipole length. Owing to poor contact with the myocardium and to the EP electrode surface area in our settings, measured LVP thresholds do not match those measured in clinical practice with standard LVP leads. However, for the study purposes, we were more interested in the PNS to LVP threshold relationship than in the LV threshold absolute value. Most importantly, we were able to locate the distance of a 2-mm-spaced dipole from the TPNE that might warrant a very high chance of PNS avoidance (Figures 4 and 5). Indeed, 100% of the patients could avoid PNS at 10 V when stimulation occurred between 15 and 25 mm from the TPNE (Figure 5). The LVP threshold was unchanged while PNS increased; the PNS to LVP safety margin increased to 45 V when a short dipole was positioned 5 mm or further from the phrenic nerve (Figure 4), fully acceptable for managing PNS long-term in clinical practice. As reported above, a 3 V safety margin has been reported as the least efficient to manage the PNS issue.8,9,12,13 The role of a short-spaced dipole is strongly supported by experimental evidence in a canine model, as recently reported.21,22 Wecke et al21 demonstrated that a very short-spaced dipole (r1 mm in length), obtained by pacing between adjacent segments of networked electrodes, can achieve PNS avoidance in 100% of worst-case scenarios (phrenic nerve dissected and positioned above the LV lead). Our observations support the concept that multipolar LV leads should have a short-spaced dipole (1–2 mm in length) to ensure PNS management at follow-up. The location of a
Heart Rhythm, Vol 10, No 1, January 2013 short-spaced dipole respective to the lead tip might be different, depending on vein length and size, as PNS occurs most commonly at mid-to-distal posterior and lateral sites.8,9 Hence, because the optimal LV pacing site to ensure maximum mechanical improvement may have substantial individual variability,4,23 the placement of a short-spaced dipole should vary as well, from close to the lead tip to 20–25 mm proximal from the lead tip. A new family of leads would allow the implanting physician to keep the LV lead in the target site for CRT, despite being very close to the phrenic nerve. This would improve clinical outcome of CRT response5–7,24 and reduce the risk of significant morbidity and mortality.1,2,4–7,18
Conclusions and practical implications We believe that our study provides the rationale for the development of multielectrode LV leads with 1–2 mm bipolar electrode spacing. This lead design would maximize PNS management, both at implantation and at follow-up, while allowing individual targeting of the LV pacing site.
Study limitations Contrary to standard clinical practice, we used commercially available EP catheters. Thus, we cannot ensure that the same observations about short-spaced dipole stimulation can be transferred to the clinical practice where different leads are employed. The EP catheter had indeed an increased LV pacing threshold compared to conventional LV pacing leads in our patients, possibly owing to the different mechanical properties and to unpredictable electrode contact with the myocardium, as previously reported in an animal study.22 It is therefore possible that the extent of the benefit of stimulation by a short-spaced dipole is underestimated. Moreover, this is an acute study that cannot provide evidence of consistency of our findings in the long-term. Chronic studies will be needed to confirm the possibility of managing PNS at the selected LV pacing site once shortspaced dipole LV leads are manufactured.
Appendix Participating investigators and their affiliations are listed here in alphabetical order: Battistini P, Biffi M, Boriani G, Diemberger I, Martignani C, Ziacchi M, Institute of Cardiology, University of Bologna, Bologna, Italy; Baracca E, Pastore G, Zanon F, Zerbo F, Ospedale Santa Maria della Misericordia, Rovigo, Italy; Bertaglia E, Zoppo F, Ospedale Civile, Mirano, Italy; Padeletti L, Perrotta L, Pieragnoli P, Ricciardi G, University of Florence, Florence, Italy; De Santo T, Varbaro A, Medtronic Italia SpA, Italy; Yang Z, Medtronic Inc, Minneapolis, Minnesota.
Acknowledgments The authors thank Valeria Ricci and Giovanna Zucchi from Medtronic, Italy, for their helpful participation in data collection.
Biffi et al
Short-Spaced Dipole for Phrenic Stimulation Management in CRT
References 1. Bristow MR, Saxon LA, Boehmer J, et al, Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140–2150. 2. Cleland JG, Daubert JC, Erdmann E, et al, for the Cardiac ResynchronizationHeart Failure (CARE-HF) Study Investigators. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005;352: 1539–1549. 3. Butter C, Auricchio A, Stellbrink C, et al, on behalf of the Pacing Therapy for Chronic Heart Failure II (PATH-CHF-II) Study Group. Effect of resynchronization therapy stimulation site on the systolic function of heart failure patients. Circulation 2001;104:3026–3029. 4. Dekker AL, Phelps B, Dijkman B, et al, Epicardial left ventricular lead placement for cardiac resynchronization therapy: optimal pace site selection with pressurevolume loops. J Thorac Cardiovasc Surg 2004;127:1641–1647. 5. Ypenburg C, van Bommel RJ, Delgado V, et al, Optimal left ventricular lead position predicts reverse remodeling and survival after cardiac resynchronization Therapy. J Am Coll Cardiol 2008;52:1402–1409. 6. Ypenburg C, van Bommel RJ, Borleffs CJW, et al, Long-term prognosis after cardiac resynchronization therapy is related to the extent of left ventricular reverse remodeling at midterm follow-up. J Am Coll Cardiol 2009;53:483–490. 7. Khan FZ, Virdee MS, Palmer CR, et al, Targeted left ventricular lead placement to guide cardiac resynchronization therapy. The TARGET study: a randomized, controlled trial. J Am Coll Cardiol 2012;59:1509–1518. 8. Biffi M, Moschini C, Bertini M, et al, Phrenic stimulation: a challenge for cardiac resynchronization therapy. Circ Arrhythm Electrophysiol 2009;2:402–410. 9. Seifert M, Schau T, Moeller V, Neuss M, Meyhoefer J, Butter C. Influence of pacing configurations, body mass index, and position of coronary sinus lead on frequency of phrenic nerve stimulation and pacing thresholds under cardiac resynchronization therapy. Europace 2010;12:961–967. 10. Gurevitz O, Nof E, Carasso S, et al, Programmable multiple pacing configurations help to overcome high left ventricular pacing thresholds and avoid phrenic nerve stimulation. Pacing Clin Electrophysiol 2005;28:1255–1259. 11. Champagne J, Healey JS, Krahn AD, et al, ELECTION Investigators. The effect of electronic repositioning on left ventricular pacing and phrenic nerve stimulation. Europace 2011;13:409–415. 12. Biffi M, Bertini M, Ziacchi M, et al, Management of phrenic stimulation in CRT patients over the long term: still an unmet need? Pacing Clin Electrophysiol 2011;34:1201–1208.
45
13. Jastrzebski M, Bacior B, Wojciechowska W, Czarnecka W. Left ventricular lead implantation at a phrenic stimulation site is safe and effective. Europace 2011;13: 520–525. 14. Biffi M, Bertini M, Saporito D, et al, Automatic management of left ventricular stimulation: hints for technologic improvement. Pacing Clin Electrophysiol 2009;32:346–353. 15. Biffi M, Boriani G. Phrenic stimulation management in CRT patients: are we there yet?. Curr Opin Cardiol 2011;26:12–16. 16. Sperzel J, D¨anschel W, Gutleben KJ, et al, First prospective, multi-centre clinical experience with a novel left ventricular quadripolar lead. Europace 2012;14: 365–372. 17. Klein N, Klein M, Weglage H, et al, Clinical efficacy of left ventricular pacing vector programmability in cardiac resynchronization therapy defibrillator patients for management of phrenic nerve stimulation and/or elevated left ventricular pacing thresholds: insights from the Efface Phrenic Stim study. Europace 2012;14:826–832. 18. Poole JE, Gleva MJ, Mela T, et al, for the REPLACE Registry Investigators. Complication rates associated with pacemaker or implantable cardioverterdefibrillator generator replacements and upgrade procedures. Circulation 2010;122:1553–1561. 19. Kirubakaran S, Rinaldi CA. Phrenic nerve stimulation with the quadripolar left ventricular lead not overcome by “electronic repositioning.” Europace 2012;14: 608–609. 20. Parahuleva M, Chasan R, Soydan N, et al, Quadripolar left ventricular lead in a patient with CRT-D does not overcome phrenic nerve stimulation. Clin Med Insights Cardiol 2011;5:45–47. 21. Wecke L, van Hunnik A, Thompson T, et al, Networked multi-electrode left ventricular pacing lead for avoidance of phrenic nerve stimulation in a canine model. Heart Rhythm 2012;9:789–795. 22. Biffi M, Foerster L, Eastman W, et al, Effect of bipolar electrode spacing on phrenic nerve stimulation and left ventricular pacing thresholds: an acute canine study. Circ Arrhythm Electrophysiol 2012;5:815–820. 23. Derval N, Steendijk P, Gula LJ, et al, Optimizing hemodynamics in heart failure patients by systematic screening of left ventricular pacing sites: the lateral left ventricular wall and the coronary sinus are rarely the best sites. J Am Coll Cardiol 2010;55:566–575. 24. Macı´as A, Gavira J-J, Castan˜o S, Alegrı´a E, Garcı´a-Bolao I. Left ventricular pacing site in cardiac resynchronization therapy: clinical follow-up and predictors of failed lateral implant. Eur J Heart Fail 2008;10:421–427.