- Email: [email protected]
Unpredictable battery depletion of St Jude Atlas II and Atlas+ II HF implantable cardioverter-defibrillators
Unpredictable battery depletion of St Jude Atlas II and Atlasⴙ II HF implantable cardioverter-defibrillators Cevher Ozcan, MD,* Jeffrey N. Rottman, MD,† E. Kevin Heist, MD, PhD, FHRS,* Mary L. Guy, RN,* Patrick T. Ellinor, MD, PhD,* Jagmeet Singh, MD, PhD,* David J. Milan, MD,* Stephan B. Danik, MD,* Conor D. Barrett, MD,* Moussa Mansour, MD, FHRS,* Jeremy N. Ruskin, MD,* Theofanie Mela, MD* From the *Cardiac Arrhythmia Service, Massachusetts General Hospital, Boston, Massachusetts; †Division of Cardiovascular Medicine, Nashville VA Medical Center and Vanderbilt University School of Medicine, Nashville, Tennessee. BACKGROUND Predictable progression to battery depletion is necessary for device management in patients with pacemakers or implantable cardioverter-defibrillators, particularly in patients who either are pacemaker dependent or have required implantable cardioverter-defibrillator therapies. OBJECTIVE To determine the incidence and characteristics of unexpected battery depletion in patients implanted with a cardiac resynchronization therapy – defibrillator (CRT-D) device. METHODS All patients with a St Jude Atlas⫹ HF or Atlas II HF CRT-D device implanted between 2004 and 2007 at the Massachusetts General Hospital and the Nashville VA Medical Center (Vanderbilt University) were studied. All patients with early generator depletion (transition of generator voltage above specified elective replacement indicator [ERI] to end of life [EOL] in less than 90 days) were evaluated further. RESULTS Eight cases (mean age 69.6 ⫾ 9 years) with abrupt battery depletion were identified among 191 patients (4.2%) implanted with a St Jude Atlas CRT-D device. The longevity of
Introduction Implantable cardioverter-defibrillator (ICD) pulse generator longevity is a significant concern for both patients and their clinicians as ICDs and cardiac resynchronization therapy with ICDs (CRT-Ds) have been used increasingly and the consequences of generator failure can be profound.1– 4 Device longevity is associated with a number of factors including the battery technology, electronic circuit efficiency, percentage of pacing, pacing mode and programmed outputs, amount of data storage, lead impedance, frequency of device therapies, and the device programming.5– 8 An important consideration is anticipating and scheduling generator replacement prior to the onset of unreliable function in Dr Mela consulted St Jude Medical and received speaker’s honoraria from Medtronic, St Jude Medical, and Boston Scientific. Address reprint requests and correspondence: Dr Theofanie Mela, MD, Cardiac Arrhythmia Service, Massachusetts General Hospital, 55 Fruit St, GRB 109, Boston, MA 02114. E-mail address: [email protected].
8 premature depletion devices was 46.4 ⫾ 10 months (median 45 months). The battery voltage in these 8 devices decreased from a mean of 2.48 ⫾ 0.03 V (above ERI) to 2.3 ⫾ 0.08 V (below ERI) over 33.3 ⫾ 23 days (range 1–59 days; median 38.5 days). One device reached EOL status within 1 day of having battery voltage above ERI and another device within 12 days. CONCLUSION The incidence of abrupt battery depletion was 4.2% in patients implanted with a St Jude Atlas CRT-D device. No common mechanism has been identified for this failure. Close monitoring of battery voltage and timely generator replacement are required in patients with these devices. KEYWORDS Pulse generator; Malfunction; Elective replacement; Defibrillator; Generator longevity; Cardiac resynchronization ABBREVIATIONS CRT-D ⫽ cardiac resynchronization therapy with ICDs; EOL ⫽ end of life; ERI ⫽ elective replacement indicator; ICD ⫽ implantable cardioverter-defibrillator (Heart Rhythm 2012;9:717–720) © 2012 Heart Rhythm Society. All rights reserved.
association with marked battery depletion (“end-of-life” [EOL] voltage). Reduction in battery life for devices exposes patients to the risks of more frequent generator replacement and the risk of procedure complications. These risks include lead damage, infection, hematoma formation, device malfunction, or death.9 –12 Patients who are either pacemaker dependent or require an appropriate ICD therapy are at risk for syncope or sudden death if they have undetected rapid battery depletion. Follow-up of device patients is therefore predicated on balancing the risks of accurate and conservative prediction of battery depletion in order to maximize generator longevity but minimize the risks of battery depletion and unreliable device function. We present a series of 8 patients who experienced rapid and unpredictable depletion of their St Jude Atlas⫹ HF or Atlas II HF CRT-D device generators. The transition from battery voltages above the defined level for elective replace-
1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2011.12.009
718
Heart Rhythm, Vol 9, No 5, May 2012
Table 1
Characteristics of patients with early battery depletion Model, St Jude Atlas
Date of implant
Longevity (mo)
Battery voltage decrease (V) (ERI at 2.45 V)
ID
Age (y)
Disease, LVEF (%)
1
72
ICMP, 20
II HF V-365
2007
34
2.46 to ⬍2.26 in 1 d
92
2
78
ICMP, 16
HF V-343
2006
47
2.55 to ⬍2.20 in 12 d
⬎99
DDD
3
74
ICMP, 25
II HF V-366
2004
40
2.46 to 2.43 in 12 d
⬎99
DDDR
4
71
ICMP, 13
HF V-340
2005
62.7
2.48 to ⬍2.20 in 28 d
⬎99
DDDR
5
80
ICMP, 27
HF V-340
2005
37.1
94
VVIR
6
70
DCM, 25
II HF V-365
2007
43
81
DDD
7
52
DCM, 11
HF V-343
2006
59
2.46 to 2.33 in 51 d 2.46 to 2.3 in 59 d 2.48 to 2.35 in 54 d
8
60
ICMP, 15
HF V-340
2004
48
2.47 to 2.33 in 49 d
Paced (%)
Mode DDDR
⬎99
DDDR
⬎99
VVIR
Output RV, LV, RA (V@ms)
Impedance RV, LV, RA (⍀)
High voltage (⍀)
[email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], no RA [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], no RA
235, 700, 415
38
270, 440, 900
41
260, 360, 410
39
325, 480, 325
40
405, 830, no RA
33
365, 830, 435
50
510, 1075, 510
56
400, 670, no RA
41
DCM ⫽ nonischemic dilated cardiomyopathy; ERI ⫽ elective replacement indicator; LV ⫽ left ventricle; LVEF ⫽ left ventricular ejection fraction; ICMP ⫽ ischemic cardiomyopathy; RA ⫽ right atrium; RV ⫽ right ventricle; VT ⫽ ventricular tachycardia.
ment to voltages consistent with EOL behavior occurred during intervals shorter than the anticipated 90-day window.
Methods All patients who underwent St Jude Atlas II HF V-366 CRT-D, Atlas II HF V-365 CRT-D, Atlas⫹ HF V-343 CRT-D, and Atlas⫹ HF V-340 CRT-D device implantation from 2004 to 2007 and were followed at the Massachusetts General Hospital and the Nashville VA Medical Center/ Vanderbilt University Arrhythmia Service were identified through a centralized device database at the Massachusetts General Hospital and the Nashville VA Medical Center/ Vanderbilt University Arrhythmia Service. Routine follow-up was performed every 3 months with a clinic visit or remote monitoring. During routine or unscheduled device interrogations, all data were evaluated to determine battery behavior. In addition to a scheduled follow-up, device interrogation, in some cases, was performed earlier for other reasons, such as the adjustment of a parameter or the detection of a suspected arrhythmia when the patient was admitted to the hospital. Of note, the Atlas II family has a vibratory patient notifier and the capability for remote monitoring. The Atlas ⫹ HF generation did not have these capabilities. For all the patients with Atlas II devices, the patient notifiers were nominally activated. Each of the patients with either abrupt or early generator depletion was included in this case series; clinical characteristics of the patients and device performance were evaluated. Data were collected from the complete records of all patients treated and followed at the Massachusetts General Hospital and the Nashville VA Medical Center/Vanderbilt University Arrhythmia Service. Data are presented as percentages or as means ⫾ SD.
Results A total of 191 patients were identified as having had a St Jude Atlas⫹ HF or Atlas II HF CRT-D device implanted between 2004 and 2007. Of these, 8 patients (4.2%) developed rapid or unpredictable battery depletion defined as a reduction in the battery voltage from above the elective replacement indicator (ERI) to EOL status within 60 days. The ERI to EOL period is expected to be 3 months. Baseline characteristics of the patients are shown in Table 1. Their mean age was 69.6 ⫾ 9 years (median 71.5 years) at the time of implantation. The most common indication for device implantation was ischemic cardiomyopathy with heart failure (6 of 8 cases). Left ventricular systolic function was severely impaired in all patients (mean left ventricular ejection fraction 19% ⫾ 6%; range 11%–27%; median 18%). All patients had New York Heart Association class III functional status prior to the initiation of device therapy. The CRT ventricular pacing percentage was 95% ⫾ 6% (81%–99%) in the DDD or VVI mode. Atrial, ventricular, and high-voltage lead impedances were within the normal range. Pacing outputs of individual leads were adjusted on the basis of lead capture threshold (Table 1). All patients underwent successful and uncomplicated generator replacement after diagnosis of this premature depletion. The total longevity of 8 patients’ battery was 46.4 ⫾ 10 months (median 45 months) from implantation to generator replacement. Examination of the battery voltage profile over time revealed a concerning trend: battery voltage decreased from a mean of 2.48 ⫾ 0.03 V (median 2.5 V) to an EOL status with a voltage of 2.3 ⫾ 0.08 V (median 2.32 V) in a mean interval of 33.3 ⫾ 23 days (range 1–59 days) with a median of 38.5 days. In each case, the time from the last
Ozcan et al
Rapid End-of-Life Transition in St Jude Defibrillators
719
Figure 1 Longevity of St Jude Atlas⫹ HF and Atlas II HF cardiac resynchronization therapy defibrillator device in 8 patients with abrupt battery depletion is demonstrated. Battery voltage curve (Volts) during follow up (months) after device implantation shows abrupt depletion in individual case.
nonalert battery voltage (above ERI) to EOL was clearly less than 90 days. Case 1: This case developed EOL battery status in 1 day following an interrogation showing battery voltage above ERI (Figure 1). St Jude Atlas II HF V-365 device interrogation revealed a battery voltage of 2.46 V during routine follow-up, with ERI specified for this device at 2.45 V. On the following day, device interrogation showed an EOL indicator with a battery voltage of 2.26 V. No ICD therapies had occurred. The longevity of the generator was 34 months. Analysis of the generator after removal showed a random component failure without a systematic problem. Of note, the patient was pacemaker-dependent with no underlying escape rhythm. Case 2: Device (Atlas⫹ HF V-343) interrogation revealed a battery voltage of less than ⬍2.20 V with EOL during routine follow-up. However, a device interrogation 12 days earlier showed a battery voltage of 2.55 V. There was a precipitous decrease in the battery voltage without spending any time formally occurring in ERI status (Figure 1). The battery lasted 47 months. Further analysis of the generator determined no specific cause for this premature depletion of the battery. Case 3: Remote device-monitoring detected that the generator (Atlas II HF V-366) had reached an ERI status (ⱕ2.45 V) with an automatic battery voltage of 2.43 V. However, 12 days earlier, interrogation of the device in clinic demonstrated a battery voltage of 2.46 V (Figure 1). The battery voltage was found to be even lower at 2.37 V 15 days later, on the day of the generator replacement. The battery lasted 40 months. No cause of this premature depletion was detected by the bench analysis of the generator. Case 4: The battery voltage was 2.48 V 28 days earlier; however, during a routine follow-up visit, Atlas HF V-340 generator was found to be in EOL with a battery voltage of 2.20 V (Figure 1) when the device was reinterrogated. Case 5: Battery depletion (Atlas HF V-340) was detected during routine interrogation when the battery voltage had been reduced from 2.46 V (above ERI) to EOL (2.33 V) within 51 days (Figure 1). The longevity of the device was 37.1 months. Analysis of the generator detected a problem with high current owing to the static RAM chip usage for storing diagnostic electrograms (EGMs). While normal static RAM should consume approximately 0.001 A in the quiescent mode, his EGM was consuming 20 A of the current in the quiescent mode. This problem doubled
the current used by the device and significantly reduced the longevity of the battery. Case 6: Battery voltage dropped to the EOL status (2.31 V) from a nonalert battery voltage of 2.46 V (above ERI) in 59 days after battery longevity of 43 months in this case (Figure 1). The battery had no time in ERI. Bench analysis of the generator (Atlas II HF V-365) showed no specific anomaly. Case 7: During routine device follow-up, battery voltage was found to be prematurely depleted by decreasing from 2.48 to 2.35 V in 54 days in this device, Atlas HF V-343 (Figure 1). The battery voltage was well above the ERI limit prior to this sudden depletion. Case 8: Similarly, unexpected battery voltage depletion occurred in this device (Atlas⫹ HF V-340). The battery voltage was above ERI at 2.47 V and decreased to 2.33 V in 49 days (Figure 1). Total longevity of the battery was 48 months. There were no stored arrhythmic events or ICD shocks or changes in pacing output when the depletion occurred in all cases. Thus far, all generators have been examined by the manufacturer in detail. Only 2 cases revealed a possible explanation for the behaviors noted. Of note, cases 1, 3, and 8 with devices with the capability of a vibratory patient notification did not report any such sensation when their devices were found at ERI/EOL. In addition, an Atlas⫹ DR V-243 ICD battery was noted to deplete abruptly from 2.6 to ⬍2.2 V in 30 days by going into EOL from the above ERI status. This generator lasted 56 months.
Discussion We present 8 cases of rapid battery depletion out of 191 or 4.2% of the St Jude Atlas family CRT-D devices followed in 2 institutions. Of particular concern was a rapid decrease in the battery voltage as the device neared EOL. Overall, these 8 devices had an average longevity of 46.4 months compared with the warranted longevity of 60 months and the expected longevity of 68 months as per the manufacturer manual. No patient received frequent ICD shocks to explain the rapid battery depletion. The percentage of ventricular pacing was high in most of the devices as is expected with many CRT devices. The most problematic aspect of device performance, however, was a rapid transition from voltages above specified ERI to EOL during a time interval that may have eluded usual device surveillance and usual scheduling of generator replacement. Remote monitoring and vibratory
720 patient notifiers were available for the Atlas II devices, but these features did not manage to bring patients to medical attention earlier. ICDs are life-saving devices, and consequently, device malfunction can have lethal consequences. As reported by Maisel et al,2 a total of 8489 ICDs out of 415,780 implanted between 1990 and 2002 were explanted owing to confirmed malfunction based on postapproval annual reports submitted to the US Food and Drug Administration by manufacturers. Battery/capacitor abnormalities and electrical issues accounted for half of the total device failures. ICD malfunction accounted for 31 patient deaths. Our report emphasizes the need for rigid premarket release testing of battery longevity to ensure better behavior in the clinical setting. Also, our cases highlight the significance of thorough postmarketing monitoring of devices. It should be noted that the only harm for the 8 reported patients was the need for early generator replacement (with its attendant risks). Had this rapid battery depletion occurred and not been detected in a pacemaker-dependent patient or a patient requiring ICD therapies, the result could have been catastrophic. Prior studies have shown variability in the ICD generator longevity among different manufacturers.6 – 8 In one series, Medtronic devices were found to have a longer life span when compared with those of Boston Scientific (Guidant) or St Jude Medical.7 Since each manufacturer has its own battery production and design, differences in battery longevity may be expected. Accurate detection of ICD battery voltage and longevity can be challenging because of its complex structure, including lithium anodes and silver vanadium oxide cathodes.8 –13 Unlike lithium-iodine pacemaker batteries, depletion of ICD battery is not a slow and steady process. In general, for Atlas⫹/Atlas HF family devices, the battery voltage remains at or near beginning-of-life (3.2 V) for some time and then gradually decreases to about 2.6 V where a plateau is typically seen. After this plateau period, the battery voltage gradually decreases (rolls off) to ERI (2.45 V) and further battery depletion could result in EOL (2.35 V). Thus, the estimation of longevity in CRT-D devices may not be precise even on the basis of the consumed and available battery voltage for each patient. Premature depletion of the battery voltage can occur and be related to the failure of any built-in components of the CRT-D systems or excessive external power drain owing to pacing or capacitor charging, including continuous pacing at high pacing outputs, lead insulation failure, and capacitor charging with recurrent aborted or delivered shocks. The latest manufacturer product performance report indicates the prevalence of premature battery depletion Atlas CRT devices at 0.12%– 0.24%. However, there was no reported abrupt or unpredictable depletion. Also, device malfunction can be a result of random component or systemic failure, including integrated circuits, resistors, capacitors, diodes, electrical interconnect, battery, and software or firmware function.
Heart Rhythm, Vol 9, No 5, May 2012 To date, no consistent explanation has been provided by the manufacturer for the behaviors that we noted. In one device (Case 1), an anode-cathode shunting (leakage current) resulted in battery failure. In another device (Case 5), the static RAM chip used for storing diagnostic EGMs was found to cause drainage of the battery and reduction in the lifetime of the device. In 2 cases, the high pacing outputs (⬎2.5 V at 0.5 ms) may have contributed to the problem. The device manual includes a recommendation for immediate replacement once high-output pacing devices reach ERI. However, we present cases where ERI was not observed before the device reached EOL. The causes of failure in the remaining 4 generators were undetermined. On the basis of the current information, we recommend the timely replacement of St Jude Atlas generators when they reach ERI (2.45 V) in patients with an underlying escape rhythm. However, we suggest replacement of the generator when the battery reaches 2.5 V (prior to ERI) in patients who are pacemaker dependent, patients who have received ICD therapies, or patients who have high pacing thresholds (⬎2.5 V at 0.5 ms). Moreover, we would recommend that a device interrogation be performed monthly after the battery voltage reaches 2.5 V. A more extensive survey of generator battery performance immediately prior to generator replacement would allow further refinement of these recommendations.
References 1. Epstein AE, Dimarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities. Heart Rhythm 2008;5:e1– e62. 2. Maisel WH, Moynahan M, Zuckerman BD, et al. Pacemaker and ICD generator malfunctions: analysis of Food and Drug Administration annual reports. JAMA 2006;295:1901–1906. 3. Hauser RG. Importance of ICD pulse generator longevity. Heart Rhythm 2009; 6:1744. 4. Ramachandra I. Impact of ICD battery longevity on need for device replacements insights from a Veterans Affairs database. Pacing Clin Electrophysiol 2010;33:314 –319. 5. Crossley GH, Fitzgerald DM. Estimating defibrillator longevity: a need for an objective comparison. Pacing Clin Electrophysiol 2007;20:1897–1991. 6. Knops P, Theuns DA, Res JC, Jordaens L. Analysis of implantable defibrillator longevity under clinical circumstances: implications for device selection. Pacing Clin Electrophysiol 2009;32:1276 –1285. 7. Schaer BA, Koller MT, Sticherling C. Longevity of implantable cardioverter defibrillators, influencing factors, and comparison to industry-projected longevity. Heart Rhythm 2009;6:1737–1743. 8. Ellinor PT, Guy ML, Ruskin JN, McGovern BA. Variability in implantable cardioverter defibrillator pulse generator longevity between manufacturers. Pacing Clin Electrophysiol 2003;26:71–75. 9. Poole JE, Gleva MJ, Mela T, et al. Complication rates associated with pacemaker or implantable cardioverter-defibrillator generator replacements and upgrade procedures: results from the REPLACE Registry. Circulation 2010;122: 1553–1561. 10. Krahn AD, Lee DS, Birnie D, et al. Predictors of short-term complications after implantable cardioverter-defibrillator replacement: results from the Ontario ICD Database. Circ Arrhythm Electrophysiol 2011;4:136 –142. 11. Hauser RG, Hayes DL, Epstein AE, et al. Multicenter experience with failed and recalled implantable cardioverter-defibrillator pulse generators. Heart Rhythm 2006;3:640–644. 12. Maisel WH. Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA 2006;295:1929 –1934. 13. Takeuchi ES, Quattrini PJ, Greatbatch W. Lithium/silver vanadium oxide batteries for implantable defibrillators. Pacing Clin Electrophysiol 1988;11:2035– 2039.
Introduction Implantable cardioverter-defibrillator (ICD) pulse generator longevity is a significant concern for both patients and their clinicians as ICDs and cardiac resynchronization therapy with ICDs (CRT-Ds) have been used increasingly and the consequences of generator failure can be profound.1– 4 Device longevity is associated with a number of factors including the battery technology, electronic circuit efficiency, percentage of pacing, pacing mode and programmed outputs, amount of data storage, lead impedance, frequency of device therapies, and the device programming.5– 8 An important consideration is anticipating and scheduling generator replacement prior to the onset of unreliable function in Dr Mela consulted St Jude Medical and received speaker’s honoraria from Medtronic, St Jude Medical, and Boston Scientific. Address reprint requests and correspondence: Dr Theofanie Mela, MD, Cardiac Arrhythmia Service, Massachusetts General Hospital, 55 Fruit St, GRB 109, Boston, MA 02114. E-mail address: [email protected].
8 premature depletion devices was 46.4 ⫾ 10 months (median 45 months). The battery voltage in these 8 devices decreased from a mean of 2.48 ⫾ 0.03 V (above ERI) to 2.3 ⫾ 0.08 V (below ERI) over 33.3 ⫾ 23 days (range 1–59 days; median 38.5 days). One device reached EOL status within 1 day of having battery voltage above ERI and another device within 12 days. CONCLUSION The incidence of abrupt battery depletion was 4.2% in patients implanted with a St Jude Atlas CRT-D device. No common mechanism has been identified for this failure. Close monitoring of battery voltage and timely generator replacement are required in patients with these devices. KEYWORDS Pulse generator; Malfunction; Elective replacement; Defibrillator; Generator longevity; Cardiac resynchronization ABBREVIATIONS CRT-D ⫽ cardiac resynchronization therapy with ICDs; EOL ⫽ end of life; ERI ⫽ elective replacement indicator; ICD ⫽ implantable cardioverter-defibrillator (Heart Rhythm 2012;9:717–720) © 2012 Heart Rhythm Society. All rights reserved.
association with marked battery depletion (“end-of-life” [EOL] voltage). Reduction in battery life for devices exposes patients to the risks of more frequent generator replacement and the risk of procedure complications. These risks include lead damage, infection, hematoma formation, device malfunction, or death.9 –12 Patients who are either pacemaker dependent or require an appropriate ICD therapy are at risk for syncope or sudden death if they have undetected rapid battery depletion. Follow-up of device patients is therefore predicated on balancing the risks of accurate and conservative prediction of battery depletion in order to maximize generator longevity but minimize the risks of battery depletion and unreliable device function. We present a series of 8 patients who experienced rapid and unpredictable depletion of their St Jude Atlas⫹ HF or Atlas II HF CRT-D device generators. The transition from battery voltages above the defined level for elective replace-
1547-5271/$ -see front matter © 2012 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2011.12.009
718
Heart Rhythm, Vol 9, No 5, May 2012
Table 1
Characteristics of patients with early battery depletion Model, St Jude Atlas
Date of implant
Longevity (mo)
Battery voltage decrease (V) (ERI at 2.45 V)
ID
Age (y)
Disease, LVEF (%)
1
72
ICMP, 20
II HF V-365
2007
34
2.46 to ⬍2.26 in 1 d
92
2
78
ICMP, 16
HF V-343
2006
47
2.55 to ⬍2.20 in 12 d
⬎99
DDD
3
74
ICMP, 25
II HF V-366
2004
40
2.46 to 2.43 in 12 d
⬎99
DDDR
4
71
ICMP, 13
HF V-340
2005
62.7
2.48 to ⬍2.20 in 28 d
⬎99
DDDR
5
80
ICMP, 27
HF V-340
2005
37.1
94
VVIR
6
70
DCM, 25
II HF V-365
2007
43
81
DDD
7
52
DCM, 11
HF V-343
2006
59
2.46 to 2.33 in 51 d 2.46 to 2.3 in 59 d 2.48 to 2.35 in 54 d
8
60
ICMP, 15
HF V-340
2004
48
2.47 to 2.33 in 49 d
Paced (%)
Mode DDDR
⬎99
DDDR
⬎99
VVIR
Output RV, LV, RA (V@ms)
Impedance RV, LV, RA (⍀)
High voltage (⍀)
[email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], no RA [email protected], [email protected] [email protected], [email protected], [email protected] [email protected], [email protected], no RA
235, 700, 415
38
270, 440, 900
41
260, 360, 410
39
325, 480, 325
40
405, 830, no RA
33
365, 830, 435
50
510, 1075, 510
56
400, 670, no RA
41
DCM ⫽ nonischemic dilated cardiomyopathy; ERI ⫽ elective replacement indicator; LV ⫽ left ventricle; LVEF ⫽ left ventricular ejection fraction; ICMP ⫽ ischemic cardiomyopathy; RA ⫽ right atrium; RV ⫽ right ventricle; VT ⫽ ventricular tachycardia.
ment to voltages consistent with EOL behavior occurred during intervals shorter than the anticipated 90-day window.
Methods All patients who underwent St Jude Atlas II HF V-366 CRT-D, Atlas II HF V-365 CRT-D, Atlas⫹ HF V-343 CRT-D, and Atlas⫹ HF V-340 CRT-D device implantation from 2004 to 2007 and were followed at the Massachusetts General Hospital and the Nashville VA Medical Center/ Vanderbilt University Arrhythmia Service were identified through a centralized device database at the Massachusetts General Hospital and the Nashville VA Medical Center/ Vanderbilt University Arrhythmia Service. Routine follow-up was performed every 3 months with a clinic visit or remote monitoring. During routine or unscheduled device interrogations, all data were evaluated to determine battery behavior. In addition to a scheduled follow-up, device interrogation, in some cases, was performed earlier for other reasons, such as the adjustment of a parameter or the detection of a suspected arrhythmia when the patient was admitted to the hospital. Of note, the Atlas II family has a vibratory patient notifier and the capability for remote monitoring. The Atlas ⫹ HF generation did not have these capabilities. For all the patients with Atlas II devices, the patient notifiers were nominally activated. Each of the patients with either abrupt or early generator depletion was included in this case series; clinical characteristics of the patients and device performance were evaluated. Data were collected from the complete records of all patients treated and followed at the Massachusetts General Hospital and the Nashville VA Medical Center/Vanderbilt University Arrhythmia Service. Data are presented as percentages or as means ⫾ SD.
Results A total of 191 patients were identified as having had a St Jude Atlas⫹ HF or Atlas II HF CRT-D device implanted between 2004 and 2007. Of these, 8 patients (4.2%) developed rapid or unpredictable battery depletion defined as a reduction in the battery voltage from above the elective replacement indicator (ERI) to EOL status within 60 days. The ERI to EOL period is expected to be 3 months. Baseline characteristics of the patients are shown in Table 1. Their mean age was 69.6 ⫾ 9 years (median 71.5 years) at the time of implantation. The most common indication for device implantation was ischemic cardiomyopathy with heart failure (6 of 8 cases). Left ventricular systolic function was severely impaired in all patients (mean left ventricular ejection fraction 19% ⫾ 6%; range 11%–27%; median 18%). All patients had New York Heart Association class III functional status prior to the initiation of device therapy. The CRT ventricular pacing percentage was 95% ⫾ 6% (81%–99%) in the DDD or VVI mode. Atrial, ventricular, and high-voltage lead impedances were within the normal range. Pacing outputs of individual leads were adjusted on the basis of lead capture threshold (Table 1). All patients underwent successful and uncomplicated generator replacement after diagnosis of this premature depletion. The total longevity of 8 patients’ battery was 46.4 ⫾ 10 months (median 45 months) from implantation to generator replacement. Examination of the battery voltage profile over time revealed a concerning trend: battery voltage decreased from a mean of 2.48 ⫾ 0.03 V (median 2.5 V) to an EOL status with a voltage of 2.3 ⫾ 0.08 V (median 2.32 V) in a mean interval of 33.3 ⫾ 23 days (range 1–59 days) with a median of 38.5 days. In each case, the time from the last
Ozcan et al
Rapid End-of-Life Transition in St Jude Defibrillators
719
Figure 1 Longevity of St Jude Atlas⫹ HF and Atlas II HF cardiac resynchronization therapy defibrillator device in 8 patients with abrupt battery depletion is demonstrated. Battery voltage curve (Volts) during follow up (months) after device implantation shows abrupt depletion in individual case.
nonalert battery voltage (above ERI) to EOL was clearly less than 90 days. Case 1: This case developed EOL battery status in 1 day following an interrogation showing battery voltage above ERI (Figure 1). St Jude Atlas II HF V-365 device interrogation revealed a battery voltage of 2.46 V during routine follow-up, with ERI specified for this device at 2.45 V. On the following day, device interrogation showed an EOL indicator with a battery voltage of 2.26 V. No ICD therapies had occurred. The longevity of the generator was 34 months. Analysis of the generator after removal showed a random component failure without a systematic problem. Of note, the patient was pacemaker-dependent with no underlying escape rhythm. Case 2: Device (Atlas⫹ HF V-343) interrogation revealed a battery voltage of less than ⬍2.20 V with EOL during routine follow-up. However, a device interrogation 12 days earlier showed a battery voltage of 2.55 V. There was a precipitous decrease in the battery voltage without spending any time formally occurring in ERI status (Figure 1). The battery lasted 47 months. Further analysis of the generator determined no specific cause for this premature depletion of the battery. Case 3: Remote device-monitoring detected that the generator (Atlas II HF V-366) had reached an ERI status (ⱕ2.45 V) with an automatic battery voltage of 2.43 V. However, 12 days earlier, interrogation of the device in clinic demonstrated a battery voltage of 2.46 V (Figure 1). The battery voltage was found to be even lower at 2.37 V 15 days later, on the day of the generator replacement. The battery lasted 40 months. No cause of this premature depletion was detected by the bench analysis of the generator. Case 4: The battery voltage was 2.48 V 28 days earlier; however, during a routine follow-up visit, Atlas HF V-340 generator was found to be in EOL with a battery voltage of 2.20 V (Figure 1) when the device was reinterrogated. Case 5: Battery depletion (Atlas HF V-340) was detected during routine interrogation when the battery voltage had been reduced from 2.46 V (above ERI) to EOL (2.33 V) within 51 days (Figure 1). The longevity of the device was 37.1 months. Analysis of the generator detected a problem with high current owing to the static RAM chip usage for storing diagnostic electrograms (EGMs). While normal static RAM should consume approximately 0.001 A in the quiescent mode, his EGM was consuming 20 A of the current in the quiescent mode. This problem doubled
the current used by the device and significantly reduced the longevity of the battery. Case 6: Battery voltage dropped to the EOL status (2.31 V) from a nonalert battery voltage of 2.46 V (above ERI) in 59 days after battery longevity of 43 months in this case (Figure 1). The battery had no time in ERI. Bench analysis of the generator (Atlas II HF V-365) showed no specific anomaly. Case 7: During routine device follow-up, battery voltage was found to be prematurely depleted by decreasing from 2.48 to 2.35 V in 54 days in this device, Atlas HF V-343 (Figure 1). The battery voltage was well above the ERI limit prior to this sudden depletion. Case 8: Similarly, unexpected battery voltage depletion occurred in this device (Atlas⫹ HF V-340). The battery voltage was above ERI at 2.47 V and decreased to 2.33 V in 49 days (Figure 1). Total longevity of the battery was 48 months. There were no stored arrhythmic events or ICD shocks or changes in pacing output when the depletion occurred in all cases. Thus far, all generators have been examined by the manufacturer in detail. Only 2 cases revealed a possible explanation for the behaviors noted. Of note, cases 1, 3, and 8 with devices with the capability of a vibratory patient notification did not report any such sensation when their devices were found at ERI/EOL. In addition, an Atlas⫹ DR V-243 ICD battery was noted to deplete abruptly from 2.6 to ⬍2.2 V in 30 days by going into EOL from the above ERI status. This generator lasted 56 months.
Discussion We present 8 cases of rapid battery depletion out of 191 or 4.2% of the St Jude Atlas family CRT-D devices followed in 2 institutions. Of particular concern was a rapid decrease in the battery voltage as the device neared EOL. Overall, these 8 devices had an average longevity of 46.4 months compared with the warranted longevity of 60 months and the expected longevity of 68 months as per the manufacturer manual. No patient received frequent ICD shocks to explain the rapid battery depletion. The percentage of ventricular pacing was high in most of the devices as is expected with many CRT devices. The most problematic aspect of device performance, however, was a rapid transition from voltages above specified ERI to EOL during a time interval that may have eluded usual device surveillance and usual scheduling of generator replacement. Remote monitoring and vibratory
720 patient notifiers were available for the Atlas II devices, but these features did not manage to bring patients to medical attention earlier. ICDs are life-saving devices, and consequently, device malfunction can have lethal consequences. As reported by Maisel et al,2 a total of 8489 ICDs out of 415,780 implanted between 1990 and 2002 were explanted owing to confirmed malfunction based on postapproval annual reports submitted to the US Food and Drug Administration by manufacturers. Battery/capacitor abnormalities and electrical issues accounted for half of the total device failures. ICD malfunction accounted for 31 patient deaths. Our report emphasizes the need for rigid premarket release testing of battery longevity to ensure better behavior in the clinical setting. Also, our cases highlight the significance of thorough postmarketing monitoring of devices. It should be noted that the only harm for the 8 reported patients was the need for early generator replacement (with its attendant risks). Had this rapid battery depletion occurred and not been detected in a pacemaker-dependent patient or a patient requiring ICD therapies, the result could have been catastrophic. Prior studies have shown variability in the ICD generator longevity among different manufacturers.6 – 8 In one series, Medtronic devices were found to have a longer life span when compared with those of Boston Scientific (Guidant) or St Jude Medical.7 Since each manufacturer has its own battery production and design, differences in battery longevity may be expected. Accurate detection of ICD battery voltage and longevity can be challenging because of its complex structure, including lithium anodes and silver vanadium oxide cathodes.8 –13 Unlike lithium-iodine pacemaker batteries, depletion of ICD battery is not a slow and steady process. In general, for Atlas⫹/Atlas HF family devices, the battery voltage remains at or near beginning-of-life (3.2 V) for some time and then gradually decreases to about 2.6 V where a plateau is typically seen. After this plateau period, the battery voltage gradually decreases (rolls off) to ERI (2.45 V) and further battery depletion could result in EOL (2.35 V). Thus, the estimation of longevity in CRT-D devices may not be precise even on the basis of the consumed and available battery voltage for each patient. Premature depletion of the battery voltage can occur and be related to the failure of any built-in components of the CRT-D systems or excessive external power drain owing to pacing or capacitor charging, including continuous pacing at high pacing outputs, lead insulation failure, and capacitor charging with recurrent aborted or delivered shocks. The latest manufacturer product performance report indicates the prevalence of premature battery depletion Atlas CRT devices at 0.12%– 0.24%. However, there was no reported abrupt or unpredictable depletion. Also, device malfunction can be a result of random component or systemic failure, including integrated circuits, resistors, capacitors, diodes, electrical interconnect, battery, and software or firmware function.
Heart Rhythm, Vol 9, No 5, May 2012 To date, no consistent explanation has been provided by the manufacturer for the behaviors that we noted. In one device (Case 1), an anode-cathode shunting (leakage current) resulted in battery failure. In another device (Case 5), the static RAM chip used for storing diagnostic EGMs was found to cause drainage of the battery and reduction in the lifetime of the device. In 2 cases, the high pacing outputs (⬎2.5 V at 0.5 ms) may have contributed to the problem. The device manual includes a recommendation for immediate replacement once high-output pacing devices reach ERI. However, we present cases where ERI was not observed before the device reached EOL. The causes of failure in the remaining 4 generators were undetermined. On the basis of the current information, we recommend the timely replacement of St Jude Atlas generators when they reach ERI (2.45 V) in patients with an underlying escape rhythm. However, we suggest replacement of the generator when the battery reaches 2.5 V (prior to ERI) in patients who are pacemaker dependent, patients who have received ICD therapies, or patients who have high pacing thresholds (⬎2.5 V at 0.5 ms). Moreover, we would recommend that a device interrogation be performed monthly after the battery voltage reaches 2.5 V. A more extensive survey of generator battery performance immediately prior to generator replacement would allow further refinement of these recommendations.
References 1. Epstein AE, Dimarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for device-based therapy of cardiac rhythm abnormalities. Heart Rhythm 2008;5:e1– e62. 2. Maisel WH, Moynahan M, Zuckerman BD, et al. Pacemaker and ICD generator malfunctions: analysis of Food and Drug Administration annual reports. JAMA 2006;295:1901–1906. 3. Hauser RG. Importance of ICD pulse generator longevity. Heart Rhythm 2009; 6:1744. 4. Ramachandra I. Impact of ICD battery longevity on need for device replacements insights from a Veterans Affairs database. Pacing Clin Electrophysiol 2010;33:314 –319. 5. Crossley GH, Fitzgerald DM. Estimating defibrillator longevity: a need for an objective comparison. Pacing Clin Electrophysiol 2007;20:1897–1991. 6. Knops P, Theuns DA, Res JC, Jordaens L. Analysis of implantable defibrillator longevity under clinical circumstances: implications for device selection. Pacing Clin Electrophysiol 2009;32:1276 –1285. 7. Schaer BA, Koller MT, Sticherling C. Longevity of implantable cardioverter defibrillators, influencing factors, and comparison to industry-projected longevity. Heart Rhythm 2009;6:1737–1743. 8. Ellinor PT, Guy ML, Ruskin JN, McGovern BA. Variability in implantable cardioverter defibrillator pulse generator longevity between manufacturers. Pacing Clin Electrophysiol 2003;26:71–75. 9. Poole JE, Gleva MJ, Mela T, et al. Complication rates associated with pacemaker or implantable cardioverter-defibrillator generator replacements and upgrade procedures: results from the REPLACE Registry. Circulation 2010;122: 1553–1561. 10. Krahn AD, Lee DS, Birnie D, et al. Predictors of short-term complications after implantable cardioverter-defibrillator replacement: results from the Ontario ICD Database. Circ Arrhythm Electrophysiol 2011;4:136 –142. 11. Hauser RG, Hayes DL, Epstein AE, et al. Multicenter experience with failed and recalled implantable cardioverter-defibrillator pulse generators. Heart Rhythm 2006;3:640–644. 12. Maisel WH. Pacemaker and ICD generator reliability: meta-analysis of device registries. JAMA 2006;295:1929 –1934. 13. Takeuchi ES, Quattrini PJ, Greatbatch W. Lithium/silver vanadium oxide batteries for implantable defibrillators. Pacing Clin Electrophysiol 1988;11:2035– 2039.