Advertisement

Deaths caused by the failure of Riata and Riata ST implantable cardioverter-defibrillator leads

      Background

      Riata and Riata ST leads (St Jude Medical, Sylmar, CA) were recalled in 2011. These leads are prone to externalized conductors. However, it is electrical malfunctions that may result in serious adverse events, including death.

      Objective

      To assess the deaths of patients with Riata and Riata ST leads that have been reported to the Food and Drug Administration to determine whether they were due to lead malfunction. A similar analysis was performed for Quattro Secure leads (Medtronic, Inc, Minneapolis, MN).

      Methods

      In February 2012, we searched the Food and Drug Administration's Manufacturers and User Facility Device Experience database for deaths associated with Riata, Riata ST, and Quattro Secure leads. A lead-related death was a sudden or unexpected death accompanied by evidence of lead malfunction; an indeterminate death was a death that was nonsudden or evidence of lead malfunction was not provided; a not lead-related death was a death that was nonsudden and/or there was no evidence that the lead contributed to the patient's demise.

      Results

      The Manufacturers and User Facility Device Experience database search found 133 deaths; of these, 22 were caused by Riata or Riata ST lead failure and 5 were caused by Quattro Secure lead failure. Riata and Riata ST lead failure deaths were typically caused by short circuits between high-voltage components. No death was due to externalized conductors.

      Conclusions

      Riata and Riata ST implantable cardioverter-defibrillator leads are prone to high-voltage failures that have resulted in death. These failures appeared to have been caused by insulation defects that resulted in short circuiting between high-voltage components. Externalized conductors were not a factor in these deaths.

      Keywords

      Abbreviations:

      ETFE (ethylene tetrafluoroethylene), ICD (implantable cardioverter-defibrillator), MAUDE (Manufacturers and User Facility Device Experience)

      Introduction

      Riata and Riata ST implantable cardioverter-defibrillator (ICD) leads (St Jude Medical, Sylmar, CA) were removed from the market by the manufacturer in 2010 and subsequently recalled in 2011.
      • Chester K.M.
      Important product information: St Jude Medical Riata and Riata ST silicone endocardial leads.
      • Carlson M.
      • Tsung P.
      Medical device advisory.
      These multilumen silicone leads are prone to a unique inside-out insulation defect caused by the movement of the conductor cables within their lumens.
      • Carlson M.
      • Tsung P.
      Medical device advisory.
      • Hauser R.G.
      • McGriff D.
      • Kallinen Retel L.
      Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
      While much interest has been directed toward externalized conductors,
      • Duray G.Z.
      • Israel C.W.
      • Schmitt J.
      • Hohnloser S.H.
      Implantable cardioverter-defibrillator lead disintegration at the level of the tricuspid valve.
      • Krebsbach A.
      • Alhumaid F.
      • Henrickson C.A.
      • Calkins H.
      • Berger R.D.
      • Cheng A.
      Premature failure of a Riata defibrillator lead impedance change or inappropriate sensing: a case report and review of the literature.
      • Erkapic D.
      • Duray G.Z.
      • Bauernfeind T.
      • De Rosa S.
      • Hohnloser S.H.
      Insulation defects of thin high-voltage ICD leads: an underestimated problem?.
      it is an electrical malfunction, such as failure to pace or defibrillate, that poses the greatest risk to patients. These malfunctions may not be heralded by detectable signs of failure,
      • Hauser R.G.
      • McGriff D.
      • Kallinen Retel L.
      Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
      and they can result in failure to deliver life-supporting or lifesaving therapy. We therefore queried the US Food and Drug Administration's Manufacturers and User Facility Device Experience (MAUDE) database to assess deaths that may have been caused by Riata and Riata ST lead failure. To provide context, we also evaluated MAUDE database reports for deaths of patients who had Quattro Secure model 6947 ICD leads (Medtronic, Inc, Minneapolis, MN).

      Methods

      Riata and Riata ST leads

      The 8-F Riata and 7-F Riata ST leads were released in the US market in 2002 and 2005, respectively. These were removed from the market in 2010 after approximately 227,000 Riata and Riata ST leads had been sold worldwide, including 96,000 Riata and 45,000 Riata ST leads in the United States. Of the 141,000 Riata and Riata US implants, approximately 79,000 remain active.
      • Carlson M.
      • Tsung P.
      Medical device advisory.
      The multilumen dual-coil construction is illustrated in Figure 1A ; the high-voltage and 2 pace-sense cables were extruded with ethylene tetrafluoroethylene (ETFE), and the pace-sense conductor coil was strung through a tube of polytetrafluoroethylene. Single-coil Riata and Riata ST models were also available.
      Figure thumbnail gr1
      Figure 1A: Cross-section of a Riata dual-coil high-voltage implantable cardioverter-defibrillator lead. The redundant cables are coated in ethylene tetrafluoroethylene, which are shown in blue, and the central coil pace-sense conductor with stylet lumen is encased in a tube of polytetrafluoroethylene, shown in green. B: Cross-section of a Quattro Secure model 6947 dual-coil high-voltage implantable cardioverter-defibrillator lead. This lead has one ethylene tetrafluoroethylene–coated cable per lumen and the pace-sense coil, which is encased in a tube of polytetrafluoroethylene, is offset from the center of the lead.

      The Sprint Quattro Secure model 6947 lead

      The Quattro Secure model 6947 (Quattro) is an 8.6-F dual-coil multilumen silicone ICD lead that has an outer sleeve of polyurethane (Figure 1B). Since its introduction in 2001, 485,000 Quattro Secure leads have been sold worldwide, including 332,000 in the United States; of these, an estimated 231,000 are active. In contrast to Riata and Riata ST leads, the Quattro Secure has a single cable in each lumen and the pace-sense coil is offset.

      Definitions

      A lead-related death was defined as a sudden or unexpected death accompanied by evidence of a lead defect; an indeterminate death was a death that was either nonsudden or the circumstances were unclear, or evidence of a lead defect was not provided; a not lead-related death was a death that was nonsudden or expected for reasons related to the patient's health status and/or there was no evidence that the lead's performance contributed to the patient's demise. Deaths due to perforation or occurring during or as the result of a lead extraction procedure were considered indirect causes of death, and therefore not lead-related.
      Lead malfunction is failure of the device to meet its performance specifications or otherwise perform as intended.
      • Maisel W.H.
      • Hauser R.G.
      • Hammill S.C.
      • et al.
      Recommendations from the Heart Rhythm Society Task Force on Lead Performance Policies and Guidelines.
      A high-voltage conductor short circuit occurs if the silicone and the ETFE insulation break down and there is contact between the conductive metal surfaces of the high-voltage cables, a high-voltage and low-voltage conductor, or a high-voltage cable and a shocking coil or pulse generator can. Externalized conductors are cables that have breached the outer insulation and are visible outside the lead body. Inside-out abrasion is an insulation defect caused by the motion of the cables within the lead insulation lumen. A can abrasion is an insulation defect caused by contact in the pocket between the lead and the pulse generator can (housing).

      Food and Drug Administration MAUDE database

      The MAUDE database contains reports of adverse events involving medical devices reported to US manufacturers by users worldwide. In our experience,
      • Hauser R.G.
      • McGriff D.
      • Kallinen Retel L.
      Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
      • Hauser R.G.
      • Katsiyiannis W.T.
      • Gornick C.C.
      • Almquist A.K.
      • Kallinen L.M.
      Deaths and cardiovascular injuries due to device-assisted implantable cardioverter-defibrillator and pacemaker lead extraction.
      • Hauser R.G.
      • Kallinen L.M.
      • Almquist A.K.
      • Gornick C.C.
      • Katsiyiannis W.T.
      Early failure of a small-diameter high-voltage implantable cardioverter lead.
      • Hauser R.G.
      • Hayes D.L.
      • Almquist A.K.
      • et al.
      Unexpected ICD pulse generator failure due to electronic circuit damage caused by electrical overstress.
      the majority of reports originate from manufacturers and <5% are submitted by user facilities or other observers. We searched the MAUDE database on February 10, 2012, for reports of deaths suffered by patients who had a Riata, Riata ST, or Sprint Quattro Secure model 6947 ICD lead, using the simple search feature and the terms “Riata death” and “Quattro Secure death.” The results of the Quattro Secure search were refined to include model 6947 leads. Reports were excluded from the study if there was no known lead problem or allegation of lead malfunction and if a returned product analysis found no anomalies that were not caused by the explant procedure.
      Lead implant time was estimated by subtracting 3 months from the lead “age,” which is the date the lead was manufactured or shipped.

      Results

      The MAUDE database search returned 133 deaths that met our study criteria; the results are summarized in Table 1. All the MAUDE reports in this study originated from the manufacturers. The failure of a Riata or Riata ST lead caused 22 deaths, while Quattro lead failures resulted in the deaths of 5 patients. Riata lead-related deaths occurred later on average than either Riata ST or Quattro lead-related deaths (61 ± 21 months vs 36 ± 7 and 36 ± 14 months; P = .012). All Riata and 2 of the 4 Riata ST lead-related deaths involved dual-coil models. The cause was indeterminate in 38% of the deaths (50 of 133). Of the 56 not lead-related deaths, 16 were procedural—8 perforations and 8 deaths associated with lead extraction.
      Table 1Results of MAUDE database search
      RiataRiata STQuattro Secure
      n%Implanted (mo), mean ± SDn%Implanted (mo), mean ± SDn%Implanted (mo), mean ± SD
      Total deaths5410048 ± 321710027 ± 206210027 ± 30
       Lead-related183361 ± 2142436 ± 75836 ± 14
       Indeterminate203751 ± 2952433 ± 17254032 ± 34
       Not lead-related163050 ± 2785316 ± 23325223 ± 29
       Perforation36<1318<123<1
       Extraction-related5985 ± 2416642366 ± 33
      MAUDE = Manufacturers and User Facility Device Experience.

      Lead-related deaths

      Twenty-seven of the 133 deaths (20%) were caused by ICD lead failure (Table 2). Of the 71 deaths involving Riata and Riata ST leads, 31% (22 of 71) were lead-related while 5 of the 62 (8%) deaths involving Quattro leads were lead-related (P = .0022). Four of the 18 Riata (22%) and 3 of the 4 Riata ST (75%) lead-related deaths were associated with can abrasions; these 7 insulation defects resulted in short circuits and failure to deliver therapy. None of the 62 Quattro leads in this study had can abrasions or a death caused by a documented insulation defect.
      Table 2Lead-related deaths
      MDR report keyModel numberImplanted (mo)Lead problemObservations indicating lead-related death
      Riata
       12147615157094Failure to defibrillate

      Abnormal HV impedance
      Sudden death. HV alert. Device failed to deliver correct energy. Insulation defect caused electrical short.
       21496478157077OversensingSudden death. Noise on EGMs “consistent with an anomalous lead.”
       31199243157128Failure to defibrillate

      Low impedance

      Can abrasion
      Death due to untreated arrhythmia. HV cable melted. Damage consistent with can abrasion.
       42157868157161Failure to defibrillate

      Abnormal HV impedance
      Device detected VF and attempted to charge several times, but failed.
       51398532158045Failure to defibrillate

      Low HV impedance
      A VF episode occurred and the device did not deliver HV therapy and the patient expired. Interrogation revealed that the HV impedance was <10 Ω.
       62409666158047Failure to defibrillate

      Low HV impedance
      Data showed that patient experienced VF. Shock failed to defibrillate. Low lead impedance found.
       7724902158045Failure to defibrillate

      High HV impedance
      The device was unable to convert VF and the patient expired. HV lead impedance was high.
       82333426158093Failure to defibrillate

      Low HV impedance
      Patient expired because of VF arrest. Device interrogation revealed attempt to deliver HV shock but failed to defibrillate because of low HV impedance.
       92243116158058Failure to defibrillate

      Low HV impedance
      Patient in VF upon arrival of EMS. CPR unsuccessful. Device interrogation revealed it delivered 2 unsuccessful HV shocks. After the second shock, the device reported low HV lead impedance and circuit damage.
       102333285158068Low HV impedance

      Lead noise

      Failure to treat VT
      Patient collapsed during VT and the device did not deliver therapy. Interrogation revealed that the patient had received HV therapy previously because of lead noise. The HV lead impedance was zero.
       111483743158149Lead noise

      Suspected fracture
      Sudden death 1 d after noise discovered on lead in clinic. Patient had been sent home because INR was too high for surgery.
       12682121158129Abnormal HV impedanceOut–of-hospital cardiac arrest. Multiple insulation abrasions on outer insulation exposing conductors and causing a short between HV conductors, melting SVC and shock cables. Abrasions were caused by constant bending and friction.
       131648070158184Abnormal lead impedancePatient had VF and was shocked successfully. The following day, out-of-range lead impedances were observed via home monitor. The patient could not be reached and expired later that evening.
       141483744158157Failure to defibrillate Oversensing

      Can abrasion
      HV cable fractured because of can abrasion. The device charged and delivered therapy. It is believed that the lead arced to the ICD can at this time.
       1517882321581UnknownFailure to defibrillate

      High lead impedance

      Can abrasion
      Interrogation revealed attempted therapy for VF, but therapies were truncated because of a high lead impedance. Visual examination found damage consistent with friction to the can.
       162147640158083Failure to defibrillate

      HV lead issue
      The patient expired 1 h after retiring. EMS CPR was unsuccessful. Device interrogation revealed that the device was in backup mode and there was an HV lead issue. An open circuit defect occurred and the device failed to deliver therapies.
       171752002158061Failure to defibrillate

      Can abrasion
      During an HV therapy delivery, the lead arced to the can (can abrasion) and shorted the HV output circuit. It is believed that the RV lead was damaged.
       18687419158140Failure to defibrillate

      High HV impedance

      Device-induced VF
      ICD inappropriately induced the patient into VF as a result of an HV lead integrity check. Interrogation reported an HV impedance of >200 Ω. VF induced and not terminated. Patient expired.
      Riata ST
       11752323700227Failure to defibrillate

      Can abrasion

      Fracture
      Device aborted charge because of possible output circuit damage and the patient expired. Visual examination of the lead found an insulation abrasion consistent with friction with the can. The 2 RV cables were fractured at this location.
       21788235702040Failure to defibrillate

      High HV lead impedance
      The ICD was attempting therapy for VF, but therapies were truncated because of high lead impedance. Analysis found insulation abrasion at 52.2–53.4 cm from the connector pin. The RV cables were broken in this area, causing high HV lead impedance. Abrasion damage was due to friction to another device.
       31573143700038Low impedance Fracture

      Can abrasion
      ICD interrogation at hospital found that it was in alert mode due to low impedance. X-ray revealed insulation damage and lead fracture. The patient expired later that day. Analysis found can abrasion with exposed cables.
       418390187002UnknownFailure to defibrillate

      Can abrasion
      Sudden death jogging. Analysis revealed short circuit due to arcing between abraded lead and the can.
      Quattro Secure
       1714012694721Inappropriate shocks

      High impedance

      Oversensing
      Patient went to ED after being shocked. Pacing impedance > 3000 Ω. Oversensing noted on ventricular EGM causing inappropriate shocks. ICD deactivated in the ED. Patient coded while being taken to the OR and died.
       21523151694757Failure to defibrillate

      Fracture
      Cause of death was VF. Physician reported that death was possibly related to a lead malfunction. Lead analysis revealed distal conductor fracture.
       32016463694733Failure to defibrillate

      Low HV impedance

      Noise
      ICD detected VF and charged to 35 J but delivered shocks between 0.6 and 1.5 J; there was decreased impedance; the patient died. Counters showed noise and oversensing.
       42025526694733OversensingThe ICD on the day of death recorded a patient alert, VF/VT, nonsustained VT episodes, intervals, and oversensing.
       52094306694721OversensingSudden death. Lead “out of specification” based on save-to-disk file.
      CPR = cardiopulmonary resuscitation; ED = emergency department; EGM = electrogram; EMS = emergency medical services; HV = high voltage; ICD = implantable cardioverter-defibrillator; INR = international normalized ratio; MDR = Medical Device Report; OR = operation room; RV = right ventricle; SVC = superior vena cava; VF = ventricular fibrillation; VT = ventricular tachycardia.
      Abnormal high-voltage impedances were the hallmark of catastrophic Riata and Riata ST lead failure, often resulting in failure to defibrillate. Excluding can abrasions, 13 Riata and Riata ST lead-related deaths involved high-voltage conductors as signified by a high or low impedance. One patient (Table 2, Riata patient number 13) was successfully rescued by his device but died the next day after the home monitor reported out-of-range impedances. Another patient (Table 2, Riata patient number 18) reportedly was induced into ventricular fibrillation by a lead integrity check and was not rescued because the high-voltage conductor was fractured. Oversensing and lead noise was observed less frequently.
      Three Quattro lead-related deaths were associated with oversensing, indicating pace-sense conductor failure, and 2 patients (Table 2, Quattro patient numbers 2 and 3) were not rescued because of high-voltage conductor defects.

      Indeterminate lead deaths

      Overall 50 of the 133 study deaths (38%) did not satisfy the criteria for lead-related death, but there was either clinical or lead data suggesting that death may have been lead-related (Table 3). In many of these cases, the MAUDE report contained a complaint such as an allegation that the lead had malfunctioned (Table 3, Riata patient number 5), but no data were available to support the allegation, and the lead was not returned to the manufacturer for analysis. In other cases (Table 3, Riata patient number 6), a lead was returned to the manufacturer and found out of specification but no clinical information was available. Two leads (Table 3, Riata patient number 14 and Riata ST patient number 4) had externalized conductors but neither had malfunctioned. Some lead failures were associated with pulse generator abnormalities (Table 3, Riata patient numbers 11 and 18 and Riata ST patient number 1) that may have been caused by a shorted lead, but evidence of lead malfunction was lacking. One alleged fixation failure (Table 3, Quattro patient number 6) appeared to have caused tamponade, but analysis of the returned lead revealed no abnormality.
      Table 3Indeterminate deaths
      MDR report keyModelImplant (mo)Lead problem complaintReason for indeterminate cause of death
      Riata
       195785715801Failure to convertNo further information provided.
       2129954315803Sudden death

      Undersensed VF
      No evidence of lead defect provided.

      Recent implant.
       357832015802Unexpected death

      Undersensed
      No evidence of lead defect provided.

      Recent implant.

      Partial lead normal.
       41344344158014Malfunction

      Shocks
      No evidence of lead defect provided.

      Returned lead showed explant damage.
       51627419158025Allegation that lead contributed to deathNo information.
       61727190158026High impedance

      Noise

      Can abrasion
      Uncertain if death sudden, unexpected, or arrhythmic. Analysis found can abrasion.
       7731440157030High impedance at follow-upPatient found expired in automobile.
       8804327158042Low HV impedanceUncertain if death sudden, unexpected, or arrhythmic.
       91694309157251Failure to captureThe physician did not consider the incident to be caused by the device.
       102193665158253OversensingUncertain if death sudden, unexpected, or arrhythmic. Noise during shock delivery for VF.
       112028722159157HV output circuit damageAnalysis of the ICD revealed HV output damage consistent with a damaged lead. Uncertain if death sudden, unexpected, or arrhythmic.
       132361800158062Inappropriate shocksUncertain if death sudden, unexpected, or arrhythmic. Analysis found abrasion.
       142014975158071Externalized conductorsNo further information provided.
       151411505157272Failure to chargeNo evidence of lead malfunction.
       162117311158072Failure to deliver therapyNo evidence of lead malfunction.
       171830867158190Increased impedance

      Oversensing

      Noise
      Uncertain if death sudden, unexpected, or arrhythmic.
       182420589158093Failure to defibrillateUnclear if failure was pulse generator or lead.

      Lead analysis found “abrasion.”
       195974581571UnknownOversensing

      Inappropriate shocks
      Apparent lead-lead interaction.

      Uncertain if death sudden, unexpected, or arrhythmic.
       205123531581UnknownHigh HV impedanceHigh HV impedance found at implant. The physician deferred replacement and disabled the ICD. Patient died 3 d later of VF.
      Riata ST
       11600040704227No outputHV circuit defect. No evidence of lead malfunction.
       21634513700031Multiple inappropriate shocksUncertain if death sudden, unexpected, or arrhythmic.
       31932369700133Failure to detect VT/VFUncertain if death sudden, unexpected, or arrhythmic. No evidence of lead malfunction.
       42361809700259Externalized conductorsUncertain if death sudden, unexpected, or arrhythmic. No evidence of lead malfunction.
       52055037700124No allegationCan abrasion consistent with short circuit. Uncertain if death sudden, unexpected, or arrhythmic. No evidence of lead malfunction.
      Quattro Secure
       1235193669478Oversensing

      No allegation
      Uncertain if death sudden, unexpected, or arrhythmic. Oversensing detected 12 d prior to death.
       22320759694733Rapid heart rateFamily reported that patient's heart rate was 227 beats/min at the time of death and the ICD did not shock. No evidence of lead malfunction.
       32252234694721UndersensingUncertain if death sudden, unexpected, or arrhythmic.
       42160385694793Cardiac arrest

      No allegation of malfunction
      The patient received 30 appropriate shocks and the device failed to shock 14 times before the therapies were exhausted. No evidence of lead malfunction.
       563009869471Failure to sense VFAnalysis of returned lead was normal.
       616547106947<1Fixation failureDuring repositioning, the helix did not deploy. The lead was replaced. Tamponade ensued. No evidence of lead malfunction.
       763256269478No allegation of malfunctionIntermittent continuity on pin cap.
       863300669475No allegation of malfunctionReturned lead tested out of specification.
       91049739694757Oversensing

      Low impedance
      Uncertain if death sudden, unexpected, or arrhythmic.
       101049982694745High impedance

      Fracture
      Uncertain if death sudden, unexpected, or arrhythmic.
       111114185694757OversensingUncertain if death sudden, unexpected, or arrhythmic.
       121348564694721MalfunctionUncertain if death sudden, unexpected, or arrhythmic.
       13141416069472High impedanceICD found to have power-on reset and out-of-range lead impedance. Analysis of returned lead was normal. Uncertain if death sudden, unexpected, or arrhythmic.
       141524656694781No allegation of malfunctionAcute heart failure and cardiac arrest. Analysis of returned lead revealed distal fracture. No evidence of ICD malfunction.
       15164917369471OversensingUncertain if death sudden, unexpected, or arrhythmic.
       161748871694769No allegation of malfunctionReturned lead tested out of specification. Uncertain if death sudden, unexpected, or arrhythmic.
       17192522169471OversensingUncertain if death sudden, unexpected, or arrhythmic.
       181944512694721No allegation of malfunctionOversensing detected by ICD. Uncertain if death sudden, unexpected, or arrhythmic.
       191945444694721Oversensing

      Inappropriate shocks
      Oversensing detected by ICD. Uncertain if death sudden, unexpected, or arrhythmic.
       202032929694733Failure to shock reported by family memberSudden death. No evidence of lead malfunction.
       21205810569471No allegation of malfunctionUnexpected death. Recent implant. Carelink transmission indicated out-of-range impedances on day of death. Lead not returned for analysis.
       2220689866947105No allegation of malfunctionPostmortem measurements revealed oversensing and high lead impedances on day of death. Uncertain if death sudden, unexpected, or arrhythmic.
       23206898969471No allegation of malfunctionRecent implant. Fine VF recorded on day of death. ICD did not detect VF.
       24211460769471Failure of helix to deployAnalysis of lead normal. Uncertain if death sudden, unexpected, or arrhythmic.
       252219334694793High impedance

      Failure to shock
      Uncertain if death sudden, unexpected, or arrhythmic. Analysis of returned lead normal.
      Unless otherwise stated, the leads were not returned for analysis.
      HV = high voltage; ICD = implantable cardioverter-defibrillator; VF = ventricular fibrillation; VT = ventricular tachycardia.

      Not lead-related deaths

      Fifty-six deaths (42%) were not lead-related (Table 1). Sixteen of these deaths were the result of either lead extraction (n = 8) or lead perforation (n = 8). All but one extraction was performed because the lead was malfunctioning. Fourteen deaths were reported to the Food and Drug Administration without allegation of lead malfunction. Of the 12 leads returned for analysis, 1 had a minor insulation abrasion, which did not expose the conductors, and 11 were within specification.

      Discussion

      Our analysis of the MAUDE database suggests that Riata and Riata ST leads are prone to high-voltage failure, which caused the deaths of 22 patients in this study. None of these failures or deaths could be attributed to externalized conductors. The signs of failure were typical of insulation defects that resulted in short circuiting between 2 or more high-voltage components, including the high-voltage cables, shocking coils, and pulse generator can. This mechanism of failure was concisely described by the manufacturer in its MAUDE report regarding Riata patient number 1 (Table 2), namely, “Analysis found insulation abrasion resulting in an electrical short.” Electrical shorts are particularly lethal because they may occur abruptly during shock delivery, and thus failure to defibrillate may be the first and only sign of lead failure. That these defects may not be detectable by routine monitoring is underscored by a case report showing that even intraoperative testing may fail to identify a shorted Riata lead.

      Leong DP, Van Erven L. Unrecognized failure of a narrow caliber defibrillation lead: the role of defibrillation threshold testing in identifying an unprotected individual. Pacing Clin Electrophysiol. Published online February 6, 2012.

      Seven of the 22 Riata and Riata ST lead-related deaths were due to can abrasions that resulted in short circuiting between exposed high-voltage cables and the pulse generator. In these and other high-voltage failures, the pulse generator sustained electrical overstress and was unable to deliver further therapy. None of the 57 Quattro leads in this study had can abrasions. Can abrasions are not detectable by noninvasive monitoring until they produce an electrical abnormality, such as out-of-range impedance, and they may not be evident at pulse generator change.

      Leong DP, Van Erven L. Unrecognized failure of a narrow caliber defibrillation lead: the role of defibrillation threshold testing in identifying an unprotected individual. Pacing Clin Electrophysiol. Published online February 6, 2012.

      ICD follow-up clinics and monitoring personnel should scrutinize all impedance vectors, for example, coil-to-can, and they should examine coil-to-can electrograms for artifacts that may be caused by a coil-to-can short.
      Recently, we reported a study of explanted Riata and Riata ST leads that were found to have inside-out insulation abrasions, which are caused by movement of the cables within their lumens.
      • Hauser R.G.
      • McGriff D.
      • Kallinen Retel L.
      Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
      One manifestation of inside-out abrasion is externalized conductors. Two patients in our series had externalized conductors (Table 3, Riata patient number 14 and Riata ST patient number 4), but there was no evidence that they had malfunctioned or contributed to the deaths of these patients. However, a quarter of inside-out abrasions occur under the Riata or Riata ST shocking coils, where the cables may not externalize.
      • Hauser R.G.
      • McGriff D.
      • Kallinen Retel L.
      Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
      We hypothesize that inside-out insulation defects under the shocking coils are causing high-voltage cables and coils to come into contact and eventually abrade the ETFE that protects and insulates the cables (Figure 1A). Then, during shock delivery, the energy intended to defibrillate is shorted between the cables or coils. If the ETFE is only partially abraded, the first shock may defibrillate but in the process of delivering high energy, the remaining ETFE breaks down and subsequent shocks are shorted, for example, Riata patient number 13 (Table 2).
      The 5 Quattro lead-related deaths were caused by conductor fractures. Oversensing and lead fracture were also found with some Quattro indeterminate deaths (Table 3). Notably, no Quattro lead in this study had an insulation defect. Compared with Riata and Riata ST leads, Quattro leads have performed well in multicenter studies.
      • Hauser R.G.
      • Maisel W.H.
      • Friedman P.A.
      • et al.
      Longevity of Sprint Fidelis implantable cardioverter-defibrillator leads: implications for patient management.
      • Keung E.
      The VA national cardiac device surveillance experience.
      Procedural complications accounted for 16 deaths—8 perforations resulting in tamponade and 8 extraction-related deaths. We did not consider these deaths lead-related or indeterminate because they were not arrhythmic and not directly caused by lead malfunction. Nevertheless, it is sobering to observe that there were 0.6 procedural deaths in our study for every lead-related death.
      For the tens of thousands of patients worldwide who have Riata and Riata ST leads, there is an urgent need for test methods that are capable of identifying life-threatening defects and a universally accessible monitoring system that promptly alerts patients and caregivers. The manufacturer should conduct studies focused on the behavior of measurable electrical parameters prior to lead failure. Algorithms that can detect electrical abnormalities before short circuits occur are needed.
      Our study suggests that any Riata or Riata ST electrical abnormality requires immediate attention. Indeed, when a high-risk patient has a malfunctioning ICD lead, he or she should be hospitalized and monitored in a unit capable of prompt external defibrillation. The rationale for this precaution is illustrated by Riata patient number 11 (Table 2) who died suddenly at home the day after his lead was found fractured during a clinic visit. Protocols should exist in hospitals and ICD surveillance clinics that specify how patients who have malfunctioning leads should be managed.
      Our study was limited by the absence of clinical data contained in the manufacturers' reports, and many leads were not returned to the manufacturer for analysis. Thus, while we believe that the information in the MAUDE database is accurate, it is often incomplete. Consequently, we classified 50 deaths as indeterminate because important information was not reported. Postmarket surveillance in the United States relies on a passive reporting system, and thus adverse events are underreported, particularly by physicians and hospitals. Therefore, the number of deaths in this study likely underestimates the actual number of deaths that have occurred. Since the MAUDE database does not contain denominator data, we used the manufacturers' sales data to estimate the number of implants.
      The results of this study are not definitive; additional confirmatory data are needed. Our findings do not support the prophylactic replacement of normally functioning Riata or Riata ST leads. Physicians and hospitals are encouraged to return failed leads to the manufacturer for analysis together with detailed clinical information, including pulse generator diagnostic data and electrograms.

      Conclusions

      Riata and Riata ST ICD leads appear to be prone to high-voltage failures that have resulted in multiple deaths. These failures often occurred abruptly, without warning, and they appeared to be caused by insulation defects that caused short circuiting between high-voltage components. Externalized conductors did not seem to be a factor in these deaths. Studies are needed to determine whether more frequent or different monitoring may detect an impending Riata or Riata ST lead failure before it results in a catastrophic event.

      References

        • Chester K.M.
        Important product information: St Jude Medical Riata and Riata ST silicone endocardial leads.
        St Jude Medical, Sylmar, CADecember 15, 2010
        • Carlson M.
        • Tsung P.
        Medical device advisory.
        St Jude Medical, Sylmar, CANovember 28, 2011
        • Hauser R.G.
        • McGriff D.
        • Kallinen Retel L.
        Riata implantable cardioverter-defibrillator failure: analysis of explanted leads with a unique insulation defect.
        Heart Rhythm. 2012; 9: 742-749
        • Duray G.Z.
        • Israel C.W.
        • Schmitt J.
        • Hohnloser S.H.
        Implantable cardioverter-defibrillator lead disintegration at the level of the tricuspid valve.
        Heart Rhythm. 2008; 5: 1224-1225
        • Krebsbach A.
        • Alhumaid F.
        • Henrickson C.A.
        • Calkins H.
        • Berger R.D.
        • Cheng A.
        Premature failure of a Riata defibrillator lead impedance change or inappropriate sensing: a case report and review of the literature.
        J Cardiovasc Electrophysiol. 2011; 22: 1070-1072
        • Erkapic D.
        • Duray G.Z.
        • Bauernfeind T.
        • De Rosa S.
        • Hohnloser S.H.
        Insulation defects of thin high-voltage ICD leads: an underestimated problem?.
        J Cardiovasc Electrophysiol. 2011; 22: 1018-1022
        • Maisel W.H.
        • Hauser R.G.
        • Hammill S.C.
        • et al.
        Recommendations from the Heart Rhythm Society Task Force on Lead Performance Policies and Guidelines.
        Heart Rhythm. 2009; 6: 870-885
        • Hauser R.G.
        • Katsiyiannis W.T.
        • Gornick C.C.
        • Almquist A.K.
        • Kallinen L.M.
        Deaths and cardiovascular injuries due to device-assisted implantable cardioverter-defibrillator and pacemaker lead extraction.
        Europace. 2010; 12: 395-401
        • Hauser R.G.
        • Kallinen L.M.
        • Almquist A.K.
        • Gornick C.C.
        • Katsiyiannis W.T.
        Early failure of a small-diameter high-voltage implantable cardioverter lead.
        Heart Rhythm. 2007; 4: 892-896
        • Hauser R.G.
        • Hayes D.L.
        • Almquist A.K.
        • et al.
        Unexpected ICD pulse generator failure due to electronic circuit damage caused by electrical overstress.
        Pacing Clin Electrophysiol. 2001; 24: 1046-1054
      1. Leong DP, Van Erven L. Unrecognized failure of a narrow caliber defibrillation lead: the role of defibrillation threshold testing in identifying an unprotected individual. Pacing Clin Electrophysiol. Published online February 6, 2012.

        • Hauser R.G.
        • Maisel W.H.
        • Friedman P.A.
        • et al.
        Longevity of Sprint Fidelis implantable cardioverter-defibrillator leads: implications for patient management.
        Circulation. 2011; 123: 358-363
        • Keung E.
        The VA national cardiac device surveillance experience.
        (Accessed March 2, 2012)

      Linked Article

      • To the Editor—Response to “Deaths Caused by the Failure of Riata and Riata ST Implantable Cardioverter-Defibrillator Leads,” by Hauser et al
        Heart RhythmVol. 9Issue 8
        • Preview
          In the article “Deaths Caused by the Failure of Riata and Riata ST Implantable Cardioverter-Defibrillator Leads” by Hauser et al, although the authors note that there are limitations of the Manufacturer and User Facility Device Experience (MAUDE) database, they proceeded to perform analyses and draw conclusions by using methodology that is dependent on the fidelity of the database itself. The Food and Drug Administration (FDA) Web site specifically states that “MAUDE data is not intended to be used either to evaluate rates of adverse events or to compare adverse event occurrence rates across devices.” Given these known limitations, we appreciate the opportunity to provide additional perspective.
        • Full-Text
        • PDF