Steroid Elution Improves the Stimulation Threshold in an Active-Fixation Atrial Permanent Pacing Lead
A Randomized, Controlled Study
Background Prior work suggests that the addition of a steroid-eluting reservoir to a passive-fixation permanent pacemaker lead improves the stimulation threshold; however, no large randomized study has addressed this issue. Over the last several years, there has been an increase in enthusiasm for the use of active-fixation permanent pacemaker leads for various reasons in spite of the generally accepted notion that active-fixation leads have higher stimulation thresholds.
Methods and Results This multicenter, randomized, controlled study examined the difference in performance between a standard active-fixation atrial lead (Medtronic model 4058) and a steroid-eluting lead (Medtronic model 4068). Stimulation thresholds were obtained in a four-point strength-duration fashion. Evaluations of sensing and impedance were performed as well. These evaluations were performed at implantation, at weeks 1 through 4, and at weeks 6, 12, 24, and 52. Stimulation thresholds were significantly better in the steroid lead than in the nonsteroid lead at each measurement point from 1 week to 12 months. The mean 1.6-V stimulation threshold at 12 months was 0.19±0.2 ms in the steroid lead and 0.41±0.30 ms in the control lead. No acute peaking was observed with the steroid lead, whereas significant peaking was observed with the control lead. There was no difference in long-term sensing or impedance.
Conclusions Inclusion of a steroid-eluting reservoir in an active-fixation permanent pacing lead improved stimulation thresholds in both the subacute and chronic periods and therefore should extend pulse-generator longevity.
Performance of the leads is the most important factor determining the safety and longevity of permanent pacemaker systems. Two types of transvenous leads are currently in use: those with a passive-fixation mechanism and those with an active-fixation mechanism. It is generally accepted that active-fixation leads have higher stimulation thresholds than do passive-fixation leads because of the trauma associated with implantation of the active-fixation leads.1 However, over the past several years, there has been an increase in enthusiasm for the use of active-fixation permanent pacemaker leads for several reasons. Passive-fixation leads generally require placement in the atrial appendage, which may be technically difficult in patients with prior cardiac surgery or congenital heart disease. The use of active-fixation leads allows atrial mapping and placement in areas that are unsuitable for passive-fixation leads.2 Use of the lateral atrial wall for lead implantation has become routine in many centers, and some centers use active-fixation leads exclusively.3 Others have suggested that active-fixation leads reduce the incidence of dislodgments.1 Pediatric cardiologists and those physicians caring for adults with congenital heart disease have found active-fixation leads to be indispensable.4 In many cases, they allow patients who otherwise would have required epicardial systems to undergo transvenous implantation.5 6 These and other factors have supported the use of active-fixation atrial leads despite the generally accepted notion that these leads have poorer stimulation thresholds than passive-fixation leads.
Although there are no large randomized studies on this subject, several small studies have suggested that steroid elution improves the subacute and chronic thresholds of passive-fixation electrodes.7 8 9 10 11 These leads have become available generally and have widened the gap in performance between passive-fixation and active-fixation leads. Given the beneficial effect on stimulation threshold with the use of passive-fixation leads, a novel active-fixation steroid-eluting lead was developed (Medtronic model 4068). In a multicenter, randomized, controlled trial, we compared this lead (used in the atrium) with a similar active-fixation lead (Medtronic model 4058) that did not contain a steroid-eluting reservoir. This is the first report of a multicenter randomized trial comparing a steroid-eluting lead with a similar control lead.
The investigational lead (Medtronic model 4068) was a bipolar, transvenous, extendable/retractable, screw-in lead with a steroid-eluting reservoir included in the fixation apparatus. The insulation material was polyurethane. The fixation mechanism was a Bisping-type screw that was electrically active.12 A reservoir in the tip of the lead contained dexamethasone sodium phosphate (1 mg). The control lead (Medtronic model 4058) was a similarly constructed active-fixation lead without a steroid-eluting reservoir. A variety of pulse generators capable of performing pulse-width thresholds at or near 0.8, 1.6, 2.5, and 5.0 V were used.
This study was a randomized, controlled study using a 3:1 ratio between the investigational lead and the control lead. Randomization was performed at the time of implantation by use of the blinded envelope method. A center-based, variable-sized, balanced-block randomization scheme was used, which resulted in each center having similar ratios of investigational versus control leads throughout the study period. The study was approved by the Institutional Review Board of each of the participating centers.
The study group comprised 254 patients (101 [40%] female, 153 [60%] male) who were undergoing either atrial or dual-chamber pacemaker implantation. Mean age was 67.3 years. Each subject gave written informed consent in accordance with the requirements of the local Institutional Review Board. Indications for pacing were typical for a dual-chamber pacemaker population. Seventy-seven percent of the leads were implanted via the subclavian vein, 23% via the cephalic vein, and none via a jugular approach. Implantation began in February 1992. All data acquired by August 1994 are included in the present report.
Lead Function Analysis
Evaluations of lead function were performed at the time of implantation; 1, 2, 3, 4, and 6 weeks after implantation; 3 and 6 months after implantation; and every 6 months thereafter. Four-point strength-duration curves were obtained by measuring pulse-width stimulation thresholds at 5.0, 2.5, 1.6, and 0.8 V. Analysis of sensing function and impedance was also performed. Notations of complications or medical events were made.
Effect on Pulse-Generator Longevity
Estimation of the effect of programming changes on longevity is easily quantifiable in pulse generators that report battery current. The current is reported in microamperes, and if one knows the deliverable battery capacity (in ampere-hours), the expected life span of the pulse generator can be calculated easily. Using the mean data from the study, one can model the energy consumption in a typical pulse generator and estimate the effect on longevity. Two separate analyses were conducted for the study. The single-chamber analysis used a Medtronic model 8416 single-chamber, rate-responsive pacemaker. It was assumed that the pulse generator could deliver 85% of its battery capacity. It was also assumed that patients would pace 100% of the time at 70 beats per minute in the AAIR mode at the base rate. The pulse generator used for the dual-chamber model was the Medtronic model 7950 dual-chamber, rate-responsive pacemaker. For dual-chamber calculations, the same assumptions were used, substituting the DDDR mode. Ventricular output was set to 2.5 V at 0.4 ms, a setting that is achieved easily with a considerable safety margin by use of passive-fixation steroid leads.13 The voltage settings used on the atrial channel for both models were a pulse width of 3 times the stimulation threshold.14 A 600-Ω noninductive impedance source was used to simulate the patient load.
The pulse-width strength-duration curves, impedances, and P-wave amplitudes were measured at each evaluation and expressed as mean±SD. Figs 1⇓ and 2⇓ display the mean±SEM. To accommodate different follow-up schedules of the patients (ie, intensive phase versus general phase), Student’s t tests were conducted at each follow-up point. Analysis of the pulse-width threshold was evaluated at settings of 1.6 and 2.5 V. The Hochberg multiple comparison procedure was used to account for multiple inference.15 Atrial dislodgment rate was compared by use of a χ2 test.
For the steroid lead, the mean P-wave amplitude was 4.12±2.53 mV at implantation and 3.50±1.75 mV at 12 months. The control lead displayed a mean implantation P wave of 4.04±2.31 mV and was 3.78±1.33 mV at 12 months. Although there were no significant differences in the P wave at implantation or chronically, the P wave was significantly stronger in the steroid lead in weeks 1 and 2 (P≤.006). This was due to the diminution of the P wave in the nonsteroid leads and the steady P wave in the steroid group.
The mean impedance of the steroid-eluting lead at implantation was 623±125 Ω compared with 589±92 Ω for the control lead (P=.03). At 12 months, mean impedance of the steroid lead was 669±107 Ω and of the control lead was 633±132 Ω (P=NS).
Fig 1⇑ displays pulse-width thresholds over time for the steroid lead versus the nonsteroid lead at 1.6 V. At implantation, mean stimulation threshold was 0.12±0.12 ms for the steroid lead and 0.13±0.11 ms for the control lead (P=NS). At 12 months, mean threshold was 0.19±0.20 ms for the steroid lead and 0.41±0.3 ms for the control lead (P<.007). At each measurement point after implantation, the steroid lead had a significantly lower stimulation threshold than did the control lead (P<.007). Similar data for measurements at 2.5 V are displayed in Fig 2⇑. Again, there was a significantly better stimulation threshold observed in the steroid group compared with the nonsteroid group at each measurement point after implantation. Statistical comparisons between the two leads are difficult at 5.0 V and 0.8 V because at 5.0 V many of the leads did not lose capture at the minimum pulse width of the pulse generators used (0.06 ms) and at 0.8 V some of the leads failed to capture at a maximum pulse width (1.5 ms).
The most striking difference between the stimulation thresholds of the control lead and those of the investigational lead was the lack of an acute increase in threshold with the use of the steroid lead. At 2.5 V, the mean stimulation threshold of the control lead increased from 0.07±0.04 ms at implantation to 0.31±0.29 ms at 1 week, whereas the steroid lead only increased from 0.09±0.12 to 0.13±0.16 ms. Fig 3⇓ displays the mean ratio of peak stimulation threshold to the threshold at implantation by use of the 2.5-V stimulation thresholds. This ratio is significantly lower for the steroid lead. The duration of peaking was strikingly different as well. At 2.5 V, the steroid lead threshold increased to 0.13 ms at 1 week but returned to chronic values by week 2, whereas the control lead threshold remained significantly above the chronic level until week 12.
Complication rates did not differ significantly between the steroid and control leads. The dislodgment rate was 2% in the steroid group and 0% in the control group (P=.58). There were no atrial perforations.
Hypothetical Energy Savings
Using the mean data from the study, one can model the energy consumption in a typical pulse generator and estimate both energy savings and longevity improvement. The mean chronic 2.5-V pulse-width threshold was 0.11 ms in the steroid lead. This suggests that a setting of 0.30-ms volts would provide for an adequate safety margin.14 Given these voltage settings and our previously mentioned assumptions, the battery current for the steroid lead would be 8.6 μA. Since the mean threshold for the nonsteroid lead was 0.19 ms, a setting of 0.60 was chosen. The battery current at these settings was 10.2 μA. Estimated longevity based on our assumptions would be 18.0 years for the steroid group and 15.2 for the control group, a difference of 2.8 years. With use of the dual chamber model noted above, the following battery currents are obtained: 12.4 μA for the steroid atrial lead and 13.6 μA for the control atrial lead. This results in predicted longevity of 14.4 years for the steroid lead and 13.1 years for the control lead control, a difference of 1.3 years.
Since the development of transvenous electrodes, there have been a number of important advances. Various fixation mechanisms have made leads less likely to dislodge. New electrode materials have improved performance in both stimulation and sensing functions. This has allowed the use of lower energy for stimulation, and therefore pulse generator size has decreased while an acceptable longevity has been maintained. The development of active-fixation leads has facilitated the placement of transvenous lead systems in many patients in whom passive-fixation lead placement would have been difficult or impossible.
Although there are no prior large-scale, randomized, controlled trials examining the efficacy of steroid elution in permanent pacemaker leads, several small, randomized, controlled trials have demonstrated that the subacute and chronic thresholds of leads that include a steroid-eluting reservoir are significantly improved over similar leads without steroid.7 8 9 Other large, nonrandomized studies have suggested similar results.10 11 An uncontrolled study of permanent epicardial leads yielded similar results.16 17 It has been demonstrated that an adequate safety margin for capture can be achieved in the majority of patients with passive-fixation, steroid-eluting endocardial electrodes with a modest pulse width at 1.6 V, resulting in substantial improvement in longevity.17 Before this study, however, no large, multicenter, randomized trial had compared the use of steroid elution with a similar lead without steroid.
Comparison With Passive-Fixation Steroid Leads
Although there are no large studies of passive-fixation atrial leads, reasonable studies are available on the use of such leads in the ventricle. One study suggested that the typical thresholds are ≈0.21 ms at 0.8 V.13 Although these thresholds are lower than those seen in the present study, they are similar for practical purposes. Using standard safety margins,14 one would program the typical patient in the present study to receive 2.5 V and 0.3 ms or 1.6 V and 0.6 ms. To take full advantage of the passive-fixation steroid lead, the pulse generator would have to be programmed to 1.6 V and 0.4 ms. This is lower than the “comfort level” of most physicians and for practical purposes, it is rare that output settings are decreased to below 2.5 V and 0.5 ms.
This study was performed in centers with considerable experience in pacing and particularly in the placement of active-fixation electrodes. The possibility exists that the stimulation thresholds achieved will not be generalizable to the pacing population in general; however, this should not affect the relative performance of the steroid versus the nonsteroid lead. The battery longevity calculations are somewhat arbitrary but represent easily obtainable values. These values would be very much affected by the type of ventricular lead chosen and by efforts made to conserve energy through reprogramming. It is likely that much better results would be seen if a steroid lead was used and the output was programmed to twice the voltage threshold at a given pulse width. It is also likely that the use of a nonsteroid lead would result in significantly poorer longevity.
These data provide compelling evidence that the inclusion of a steroid-eluting reservoir in an active-fixation permanent pacing lead improved stimulation thresholds in both the subacute and chronic periods. Furthermore, the data suggest that if the steroid lead were substituted for the control lead and the device programmed accordingly, a significant increase in pulse generator longevity could be expected.
Investigators and Coordinators for the Medtronic 4068 Study
Yale-New Haven Hospital, New Haven, Conn: Bill Batsford, MD; Gini Elwood.
Mount Sinai Medical Center, New York, NY: Jorge L. Camunas, MD; Unsoon Shagong, RN.
Thomas Jefferson University Hospital, Philadelphia, Pa: Arnold Greenspon, MD; Michelle Romash, RN.
Baystate Medical Center, Springfield, Mass: James Kirchhoffer, MD; Laura Liucci, RN.
Brigham & Women’s Hospital, Boston, Mass: Gary Mitchell, MD; Bernadette Stanton-Mayor, RN; and Anne Spector, RN.
Children’s Hospital, Boston, Mass: Walter J. Gamble, MD.
Blake Hospital, Bradenton, Fla: Robert Batey, MD; Steve Erickson (follow-ups); Mark Sweesy; Debby Bunton; and Beth McHargue.
Johns Hopkins University, Baltimore, Md: Jeffrey Brinker, MD; Billie-Jo Kreps.
NC Baptist Hospital/Bowman Gray School of Medicine, Winston-Salem, NC: George H. Crossley, MD; Tony W. Simmons, MD; Wesley K. Haisty, Jr, MD; David M. Fitzgerald, MD; Kathleen D. O’Brien, RN; and Lisa Kiger, RN.
Baptist Montclair Medical Center, Birmingham, Ala: Russell Reeves, MD; William R. Harrison, MD; Nancy Self, RN; and Melisa Lail, RN.
Illinois Masonic Medical Center, Chicago: Marilyn Ezri, MD; Susan Donahue, RN.
Barnes Hospital, St Louis, Mo: Bruce Ferguson, MD; Dennis Fogarty, PA.
Mercy Hospital, Des Moines, Iowa: W. Ben Johnson, MD; Amy Leiserowitz, RN.
William Beaumont Hospital, Royal Oak, Mich: James Stewart, MD; Lori Bell, RN.
Seton Medical Center, Austin, Tex: Gerald A. Baugh, MD; and Corinne Wise.
St Francis Hospital, Tulsa, Okla: James Higgins, MD; Michelle Raby, RN; Martha Thompson, RN.
Methodist Hospital, Lubbock, Tex: Howard Hurd, MD; Melanie Quijada.
University of Oklahoma Hospital, Oklahoma City: Dwight Reynolds, MD; Roberta Kendall, RN.
Methodist Hospital, Houston, Tex: William Spencer, MD; Mary Freeman; and Cindy Kirkpatrick.
Mercy Hospital San Diego, Calif: Jerrold Glassman, MD; Fred McDonald.
Long Beach Memorial Hospital, Calif: John Messenger, MD; Arlene Rylaarsdam.
Central Cardiology Medical Clinic, Bakersfield, Calif: Peter Nalos, MD; Marilyn Williams.
Sacred Heart Medical Center, Spokane, Wash: David Oakes, MD; Steve Edwards, RN.
This study was sponsored and supported by a grant from Medtronic Inc, Minneapolis, Minn.
Reprint requests to George H. Crossley, MD, Cardiology Section, Bowman Gray School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157-1045. E-mail firstname.lastname@example.org.
G.H.C. is an investigator and consultant to Medtronic Inc on this and other projects. J.A.B. serves as Chairman, Food and Drug Administration Panel on Cardiac Devices. D.R. is an investigator and consultant to Medtronic Inc on other projects. L.T. and M.Z. are employees of Medtronic Inc.
↵1 A complete list of the investigators and coordinators for the study is provided in the “Appendix.”
- Received May 30, 1995.
- Accepted July 7, 1995.
- Copyright © 1995 by American Heart Association
Masterson MM, Maloney JD, Tuzcu EM, Wilkoff BL, Emre A, Vanerio G, Simmons TW, Morant VA, Castle LW. Atrial pacemaker leads compared. Cleve Clin J Med. 1990;57:433-436.
Ward DE, Clarke B, Schofield PM, Jones S, Dawkins K, Bennett D. Long term transvenous ventricular pacing in adults with congenital abnormalities of the heart and great arteries. Br Heart J. 1983;50:325-329.
Sutton R, Guneri S. The impact of steroid eluting leads on long term pacing in the atrium and ventricle. Eur J CPE. 1991;1:10-15.
Stokes KB, Kay GN. Artificial electric cardiac stimulation. In: Ellenbogen KA, Kay GN, Wilkoff BL, eds. Clinical Cardiac Pacing. Philadelphia, Pa: W.B. Saunders Co; 1995:31-32.
Dunnett CW, Tamhane AC. A step-up multiple test procedure. JASA. 1987;417:162-170.