Author + information
- Received October 3, 2017
- Revision received November 27, 2017
- Accepted November 30, 2017
- Published online May 21, 2018.
- David S. Frankel, MDa,∗ (, )
- Martin C. Burke, DOb,
- David J. Callans, MDa,
- Timothy M. Stivland, MBAc,
- Elizabeth Duffy, MSc and
- Andrew E. Epstein, MDa
- aElectrophysiology Section, Cardiovascular Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia
- bHeart Rhythm Center, CorVita Science Foundation, Chicago, Illinois
- cBoston Scientific Corporation, St. Paul, Minnesota
- ↵∗Address for correspondence
: Dr. David S. Frankel, Cardiovascular Division, Electrophysiology Section, Hospital of the University of Pennsylvania, 9 Founders Pavilion, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
Objectives This study determined whether obesity increased the risk of failed shocks and complications among subcutaneous implantable cardioverter-defibrillator (S-ICD) recipients.
Background The S-ICD is an established treatment for patients at high risk for ventricular arrhythmias. Obesity may increase the transvenous defibrillation threshold and the rate of complications.
Methods We analyzed data from the 321 patients enrolled in the S-ICD Investigational Device Exemption (IDE) study. They were categorized into 3 body mass index (BMI) groups: <25 kg/m2 (underweight and normal), 25 to 30 kg/m2 (overweight), and >30 kg/m2 (obese). Patients underwent implantation followed by defibrillation testing at 65 J. Chronic defibrillation testing was repeated >150 days post implantation in a subset of patients.
Results Seventy-nine patients had BMI <25 kg/m2, 105 had BMI 25 to 30 kg/m2, and 137 had BMI >30 kg/m2. A failed first shock of any kind occurred in 52 patients, including 41 patients during implant defibrillation testing, 11 patients during chronic defibrillation testing, and 5 patients during spontaneous ventricular arrhythmias. The rate of failed first shocks during implant defibrillation testing increased across BMI categories (5.1% among underweight and normal patients, 13.3% among overweight patients, and 16.9% among obese patients; p = 0.04). Among obese patients, shock impedance was higher during failed first shocks than successful first shocks (103.5 ohms vs. 84.6 ohms; p = 0.001). The rate of failed first shocks during chronic defibrillation testing and spontaneous ventricular arrhythmias did not significantly differ among BMI groups, nor did the rate of complications. Of the 8 underweight patients, there were no failed shocks or complications.
Conclusions Obese patients are at increased risk of failed first S-ICD shocks during defibrillation testing. Whether this can be overcome by optimal implantation techniques remains unknown. Rates of complications were not increased among obese patients.
The subcutaneous implantable cardioverter-defibrillator (S-ICD) is an established treatment for patients at high risk of ventricular arrhythmias (1,2). Obesity has been shown to increase the transvenous defibrillation threshold (DFT) in an experimental model (3). Obesity also increases risk of complications across a variety of invasive procedures (4–6). In addition, underweight patients who receive S-ICDs could be at increased risk for discomfort or erosion from the lateral device pocket. We hypothesized that obesity would increase the rate of failed appropriate shocks and that extremes of weight (underweight or obese) would increase the risk of complications among those undergoing S-ICD implantation.
Study population and defibrillation testing
Patients who underwent S-ICD implantation during the Investigational Drug Exemption (IDE) trial (NCT01064076) were studied (2). This prospective, nonrandomized, multicenter trial included adults who met standard ICD indications, without need for pacing or documented pace-terminable ventricular tachycardia. The S-ICD was implanted using the previously described standard technique, after which acute defibrillation testing was performed. Ventricular fibrillation (VF) was induced with 50-Hz transthoracic pacing. Detection was performed automatically by the device, after which an initial 65-J shock was delivered. The primary effectiveness endpoint was conversion of induced VF, in which a successful test required 2 consecutive VF conversions at 65 J in either shock vector, within a maximum of 4 VF conversion attempts using the same polarity. Conversion success on the first shock was analyzed in the present study. Chronic defibrillation testing was performed in one-quarter of patients, at least 150 days following implantation. Again, VF was induced, and an initial 65-J shock was delivered.
Clinical follow-up and endpoints
After implantation, all shocks were programmed to 80 J. Patients were examined 1, 3, and 6 months following S-ICD implantation, and every 6 months thereafter until study completion. In addition, patients were evaluated if they experienced spontaneous shocks or a change in clinical status. The S-ICD was interrogated at all study visits. Any failed first shock was defined as a failure of the first shock to terminate VF or ventricular tachycardia during acute defibrillation testing (65 J), chronic defibrillation testing (65 J), or spontaneous ventricular arrhythmia (80 J). Impedances were assessed during acute defibrillation testing. Complications were grouped into 3 categories: infection-type complications (erosion, prolonged healing, superficial and system infection), suboptimal positioning-type complications (device discomfort or suboptimal generator and/or lead position determined by the operator as requiring revision), and inappropriate shocks.
Body mass index (BMI) was calculated as the body mass divided by the square of the body height (kg/m2). Patients were assigned into BMI categories as defined by the Centers for Disease Control and Prevention, as follows: underweight (BMI <18.5 kg/m2), normal (18.5 to 24.9 kg/m2), overweight (25.0 to 29.9 kg/m2), and obese (≥30.0 kg/m2). Because of the small number of underweight patients, they were combined with normal patients for primary analyses. However, in subsidiary analyses, underweight patients were analyzed as a distinct group. Baseline characteristics, failed first shocks, and complications were compared across BMI categories. Continuous variables were expressed as mean ± SD, and categorical variables were expressed as frequencies with percentages. The 1-way analysis of variance F-test was used to compare continuous variables, and Pearson’s chi-square test or the Fisher exact test were used to compare categorical variables. We considered p values ≤0.05 to indicate statistical significance. Logistic regression was used to identify variables associated with failed first shock and complications. Univariate analysis for each model was performed. Variables with p < 0.10 in univariate modeling were candidates for inclusion in the multivariate models. Backwards selection with p ≤ 0.05 stay criteria was used to determine the final multivariate models. Analyses were performed using SAS software version 9.4 (SAS Institute, Cary, North Carolina).
Body mass index and other baseline characteristics
Of 321 patients implanted with the S-ICD, 8 (2.5%) were underweight, 71 (22.1%) were normal weight, 105 (32.7%) were overweight, and 137 (42.7%) were obese (Figure 1). Overweight and obese patients were older and more frequently men than normal and underweight patients (Table 1). Overweight and obese patients had higher mean creatinine and higher rates of medical comorbidities, including diabetes, hypertension, atrial fibrillation, and heart failure. Overweight and obese patients more frequently took diuretics, anticoagulants, and antidiabetic medications.
Failed first shocks
A failed first shock of any kind occurred in 52 patients, including 41 patients during implantation defibrillation testing, 11 patients during chronic defibrillation testing, and 5 patients during spontaneous ventricular arrhythmia episodes. The rate of failed first shocks during implantation defibrillation testing increased across BMI categories (5.1% among normal and underweight patients, 13.3% among overweight patients, and 16.9% among obese patients; p = 0.04) (Table 2). Among obese patients undergoing acute defibrillation testing, impedance was higher during failed first shocks than successful first shocks (103.5 ohms vs. 84.6 ohms; p = 0.001) (Figure 2). No such difference was observed among normal and overweight subjects. Shock impedance was highly consistent between first and second shocks (R2 = 0.98) (Online Figure 1). Of the 41 patients in whom the first shock failed during acute defibrillation testing, 24 completed the protocol successfully, with subsequent inductions without lead or generator repositioning. Another 2 patients (both obese) remained implanted without an attempt to reposition the lead or generator, despite failure of the defibrillation testing protocol. Repositioning of the lead or generator was performed in 15 patients. Eight (3 overweight and 5 obese) of these 15 patients subsequently completed the defibrillation testing protocol successfully (with a decrease in mean impedance from 81 to 77 ohms), whereas the other 7 patients did not successfully complete the protocol (with decrease in mean impedance from 107 to 92 ohms). Of these 7 patients, the device was removed in 6 (all obese) and left in place in 1, after 2 successful conversions at 70 J. Of the 41 patients with a failed first shock during acute defibrillation testing, 4 went on to have spontaneous ventricular arrhythmia episodes. Three of these episodes were terminated with the first 80-J shock. The remaining episode, which occurred in a normal weight patient, was terminated with the fifth shock.
Chronic defibrillation testing was performed in 80 patients. Rates of failed first shock during chronic defibrillation testing did not significantly differ by BMI group (p = 0.40). Of the 11 patients with a failed first shock during chronic defibrillation testing, 4 had experienced failure of the first shock during acute defibrillation testing.
Survival at 720 days was similar among the groups (96.1% among normal and underweight patients, 95.8% among overweight patients, and 98.4% among obese patients; log-rank p = 0.90). Eleven spontaneous ventricular arrhythmia episodes occurred among 4 normal and underweight patients, 16 episodes among 8 overweight patients, and 24 episodes among 13 obese patients. The rate of failure of the first 80-J shock to terminate spontaneous ventricular arrhythmias did not differ among the BMI groups (p = 0.60). Of the 5 patients with a failed first 80-J shock during spontaneous ventricular arrhythmias, only 1 had a failed first 65-J shock during acute defibrillation testing, as described previously. Without lead or generator repositioning, this normal weight patient was successfully converted during the next 2 VF episodes with 65-J shocks.
The rate of failed first shocks of any kind increased across BMI categories (6.3% among normal and underweight patients, 18.1% among overweight patients, and 20.6% among obese patients; p = 0.02). There were no failed first shocks of any kind among the 8 underweight patients. In univariate testing, male sex, younger age, taller height, higher weight, higher BMI, and primary cardiac disease other than ischemic cardiomyopathy were all associated with an increased risk of any first shock failing (Table 3). Because of the correlation with BMI, height and weight were not entered into the multivariate model. In the multivariate model, male sex, primary cardiac disease other than ischemic cardiomyopathy, and higher BMI remained significantly associated with failed first shocks. Using normal and underweight as the reference, the odds ratio for a failed first shock among overweight patients was 3.65 (95% confidence interval: 1.46 to 9.14; p = 0.006), and among obese patients, the odds ratio was 3.72 (95% confidence interval: 1.55 to 8.94; p = 0.003).
The rate of infectious type complications did not significantly differ among BMI groups (5.1% among normal and underweight patients, 2.9% among overweight patients, and 2.9% among obese patients; p = 0.70) (Table 4). The rate of suboptimal positioning complications did not differ among BMI groups (3.8% for normal and underweight patients, 1.9% for overweight patients, and 5.1% for obese patients; p = 0.40). Similarly, the rate of inappropriate shocks did not differ among BMI groups (3.8% among normal and underweight patients; 2.9% among overweight patients, and 3.9% among obese patients; p = 0.90). No complications occurred among the 8 underweight patients. Logistic regression did not identify any significant univariate predictors of infectious-type or suboptimal positioning-type complications, including BMI. History of atrial fibrillation was associated with an increased risk of inappropriate shocks (odds ratio: 3.9; 95% confidence interval: 1.07 to 14.52; p = 0.04).
Although the overall efficacy of the S-ICD was high and comparable to transvenous ICDs, we found the rate of failed first shocks during acute defibrillation testing increased with the BMI (7–9). In contrast to other invasive procedures, we did not find an increased rate of complications in obese patients. Despite concerns about the large size of the S-ICD generator, we did not find underweight patients to be at increased risk of infection, erosion or discomfort, although the number of such patients was small.
The mechanism responsible for the higher rate of failed first shocks among obese patients requires further study. It is noteworthy that during acute defibrillation testing, obese patients in whom first shocks failed had higher impedance than obese patients who experienced successful first shocks. Using a computer model, Heist et al. (10) found that adipose tissue between the generator and ribs or the lead and sternum increased both the impedance and DFT. For example, 10 mm of underlying subcutaneous fat between the lead and sternum increased the impedance from 83 to 144 ohms and the DFT from 29 to 95 J, which is above the maximum device output. Repositioning the lead more deeply, just overlying the fascia, was reported to lower the DFT in an obese patient in whom conversion testing previously failed (11). Thus, it is possible, that with an optimal implantation technique, including minimizing subcutaneous fat between the lead and sternum, as well as the generator and ribs, the rate of failed first shocks would no longer be increased among obese patients. A significant learning curve was described for S-ICD implantation, in which the rate of complications decreased with operator experience (12). Unfortunately, as is evident from the significant overlap shown in Figure 2, predicting success using impedance alone is not possible.
Obese patients have higher rates of wound infection following many types of surgery, as do patients with diabetes (6,13). Despite higher BMI and greater prevalence of diabetes, obese patients did not have a higher rate of infectious complications after S-ICD implantation in the IDE study. Nevertheless, with an overall rate of infection-type complications of 3.4%, room remains for improvement. Extending the skin preparation for the device pocket more posteriorly was found to reduce infection risk (14). The role of submuscular device pockets and antibiotic-eluting pouches should be explored. Despite concerns of discomfort and erosion in the lateral device pocket among underweight patients, we found no such complications.
There were only 8 underweight patients included in the IDE study; thus, conclusions regarding this population were limited by power. Because of multiple statistical comparisons, p values should be interpreted with caution. By testing exclusively at 65 J, we determined probabilities of successful defibrillation; however, we did not determine actual defibrillation thresholds. Because of the small number of patients who underwent chronic defibrillation testing and who experienced spontaneous ventricular arrhythmias, the study might have been underpowered to detect differences in rates of failed first shocks in these settings. Quality of life was not formally assessed as part of the IDE study; thus, patient comfort and satisfaction regarding cosmetic result could not be assessed. Although obesity remained independently associated with failed first shocks in our multivariate model, it is possible that this association was confounded by unmeasured variables, such as right ventricular dimensions. Most importantly, the mechanism of increased failed first shocks among obese patients could not be definitively determined by the present study. The implications are important. If higher DFTs can be overcome by optimal implantation technique, then attention should be focused on such technique in obese patients. Alternatively, if DFTs remain elevated despite optimal implantation technique, this should be considered when selecting defibrillator type for obese patients. Because of the lack of transvenous ICD control subjects, we cannot conclude which ICD is more effective for obese patients. The results of the ATLAS S-ICD (Avoid Transvenous Leads in Appropriate Subjects; NCT02881255) and PRAETORIAN (Prospective, Randomized Comparison of Subcutaneous and Transvenous Implantable Cardioverter Defibrillator Therapy; NCT01296022) trials, which are both randomizing patients to subcutaneous ICDs versus transvenous ICDs, are awaited in this regard.
Obese patients are at increased risk for failed first shocks during acute defibrillation testing from the S-ICD. Whether this can be overcome by optimal implantation technique requires further study. Rates of complications are not increased in obese or underweight patients.
COMPETENCY IN MEDICAL KNOWLEDGE: We found an increased rate of failed first S-ICD shocks during acute defibrillation testing among overweight (odds ratio: 3.65) and obese (odds ratio: 3.72) patients. Rates of other complications did not significantly differ.
TRANSLATIONAL OUTLOOK: Further study is needed to determine whether this can be overcome by optimal generator and lead positioning, or whether defibrillation thresholds are higher in obese patients, regardless of implantation technique.
This work was supported by the Koegel Family Fund.
Dr. Frankel has received lecture honoraria from Boston Scientific and Medtronic. Dr. Burke has received lecture honoraria from Boston Scientific, Medtronic, and Abbott; and has received research support from Boston Scientific, Cameron Health, AJ Medical, Inc., Medtronic, and Abbott. Dr. Callans has received fees as a consultant and speaker for Boston Scientific. Mr. Stivland and Ms. Duffy are employees of Boston Scientific. Dr. Epstein has received honoraria for committee membership and research support from Boston Scientific, Medtronic, and Abbott.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institution and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- body mass index
- defibrillation threshold
- subcutaneous implantable cardioverter-defibrillator
- ventricular fibrillation
- Received October 3, 2017.
- Revision received November 27, 2017.
- Accepted November 30, 2017.
- 2018 American College of Cardiology Foundation
- Burke M.C.,
- Gold M.R.,
- Knight B.P.,
- et al.
- Weiss R.,
- Knight B.P.,
- Gold M.R.,
- et al.
- Thornqvist C.,
- Gislason G.H.,
- Kober L.,
- Jensen P.F.,
- Torp-Pedersen C.,
- Andersson C.
- Heist E.K.,
- Belalcazar A.,
- Stahl W.,
- Brouwer T.F.,
- Knops R.E.
- Brouwer T.F.,
- Driessen A.H.G.,
- Olde Nordkamp L.R.A.,
- et al.
- Martin E.T.,
- Kaye K.S.,
- Knott C.,
- et al.
- Brouwer T.F.,
- Miller M.A.,
- Quast A.B.,
- et al.