https://orcid.org/0000-0003-0306-7764
Abstract
The traditional technique for subcutaneous implantable cardioverter-defibrillator (S-ICD) implantation involves three incisions and a subcutaneous pocket. Recently, a two-incision and intermuscular (IM) technique has been adopted. The PRAETORIAN score is a chest radiograph-based tool that predicts S-ICD conversion testing. We assessed whether the S-ICD implantation technique affects optimal position of the defibrillation system according to the PRAETORIAN score.
We analysed consecutive patients undergoing S-ICD implantation. The χ2 test and regression analysis were used to determine the association between the PRAETORIAN score and implantation technique. Two hundred and thirteen patients were enrolled. The S-ICD generator was positioned in an IM pocket in 174 patients (81.7%) and the two-incision approach was adopted in 199 (93.4%). According to the PRAETORIAN score, the risk of conversion failure was classified as low in 198 patients (93.0%), intermediate in 13 (6.1%), and high in 2 (0.9%). Patients undergoing the two-incision and IM technique were more likely to have a low (<90) PRAETORIAN score than those undergoing the three-incision and subcutaneous technique (two-incision: 94.0% vs. three-incision: 78.6%; P = 0.004 and IM: 96.0% vs. subcutaneous: 79.5%; P = 0.001). Intermuscular plus two-incision technique was associated with a low-risk PRAETORIAN score (hazard ratio 3.76; 95% confidence interval 1.01–14.02; P = 0.04). Shock impedance was lower in PRAETORIAN low-risk patients than in intermediate-/high-risk categories (66 vs. 96 Ohm; P = 0.001). The PRAETORIAN score did not predict shock failure at 65 J.
In this cohort of S-ICD recipients, combining the two-incision technique and IM generator implantation yielded the lowest PRAETORIAN score values, indicating optimal defibrillation system position.
http://clinicaltrials.gov/ Identifier: NCT02275637.
Subcutaneous implantable cardioverter-defibrillator implantation technique affects optimal chest-radiograph position of the defibrillation system.
Intermuscular (IM) pulse generator implantation combined with the two-incision technique is associated with a low PRAETORIAN score (<90).
The two-incision plus IM implantation technique yields lower shock impedance than any other implantation approaches.
Introduction
The subcutaneous implantable cardioverter-defibrillator (S-ICD) is a proven alternative to the traditional implantable cardioverter-defibrillator (ICD) and is envisioned to reduce transvenous lead-related complications.1–3
Optimal implantation of the S-ICD requires minimizing the amount of adipose tissue between the coil and the sternum and between the generator and the thorax.4,5 Moreover, the pulse generator (PG) needs to be properly positioned along the anterior–posterior (AP) axis of the thorax. The PRAETORIAN score is a chest radiograph-based method that assesses these determinants of the defibrillation threshold (DFT) and has been retrospectively validated to predict the probability of successful S-ICD conversion testing.6 The prospective validation of the score is ongoing in the PRAETORIAN-DFT trial.7
The traditional S-ICD implantation technique, which involves three incisions and insertion of the PG under the subcutaneous tissue, has significantly changed over time. A new technique that uses two incisions and an intermuscular (IM) pocket for the PG between the serratus anterior and the latissimus dorsi muscles has been introduced and is now widely adopted.8,9 With this technique, the superior parasternal incision is omitted, thus preventing potential skin erosions/infections and improving the cosmetic result. Moreover, the deeper position of the PG may prevent pocket complications and allows for more posterior (dorsal) placement of the device. This results in less fat interposition between the PG and the chest, potentially reducing shock impedance and DFT.5,10
In this study, we aimed to assess whether the S-ICD implantation technique affects optimal chest radiograph position of the defibrillation system according to the PRAETORIAN score.
Methods
Study population
Patients from 13 referral centres participating in the Rhythm Detect Registry, who underwent implantation of an S-ICD (Boston Scientific Inc., Natick, MA, USA) between January 2013 and July 2018, with available post-implantation posterior–anterior (PA) and lateral chest radiographs were included in this study. The Institutional Review Boards approved the study and all patients provided written informed consent for data storage and analysis. Baseline assessment comprised collection of demographic data and medical history, clinical examination, chest radiography, 12-lead electrocardiogram (ECG), estimation of New York Heart Association functional class, and echocardiographic evaluation.
Subcutaneous implantable cardioverter-defibrillator implantation
Before implantation, S-ICD eligibility was assessed through surface ECG screening by means of a dedicated ECG morphology tool or an automatic screening tool.11 In accordance with the first technique described, implantation was performed through three incisions: one on the left-lateral chest for the PG pocket, and two parasternal for lead tunnelling. The PG was positioned in a subcutaneous pocket over the fifth intercostal space between the mid and the anterior axillary lines. Newer techniques include a two-incision approach, which avoids the third superior parasternal incision by tunnelling the defibrillation lead through a peel-away sheath introducer,8,9,12 and IM generator implantation between the serratus anterior and the latissimus dorsi muscles (Figure 1). Physician preference and patient characteristics determined the implantation technique used.
(Left panel) With the IM technique, the PG is positioned at the level of the VII and VIII costal arches, dorsally to the anterior margin of the latissimus dorsi muscle, in an intermuscular pocket between the serratus anterior and the latissimus dorsi muscles. The pocket incision is made obliquely, following Langer’s lines, over the V–VI intercostal space. (Right panel) The latissimus dorsi muscle delimits two different pockets: subcutaneous and intermuscular. IM, intermuscular; PG, pulse generator.
Chest radiograph analysis
Post-operative AP and lateral chest radiographs were analysed. Quality was judged adequate if the complete coil and generator were visualizable and angulation of the views was minimal. The PRAETORIAN score was calculated according to a three-step approach.6 In brief, the Step 1 determines the amount of sub-coil fat by assessing the thickness of the adipose tissue between the coil and the sternum or ribs by using the coil width as a reference (Figure 2A and B). The second step determines whether the generator is positioned on or posterior to the midline (Figure 2C and D). The Step 3 determines the amount of sub-generator fat by using the generator width as a reference (Figure 2E and F). This results in a final PRAETORIAN score, indicating the risk of conversion failure. Patients with a body mass index (BMI) of ≤25 kg/m2 are rewarded by subtracting 40 points in the case of a score of ≥90.
Chest radiograph analysis according to the PRAETORIAN score. The amount of sub-coil fat on the lateral chest radiograph view affects the PRAETORIAN score, shock impedance and DFT. In (A), the thickness of the adipose tissue is ≤1 coil width. This corresponds to 30 points in the score system (Step 1 of the PRAETORIAN score) and low predicted risk of shock failure. In (B), the thickness of fatty tissue between the coil and the sternum is >3 coil widths. In this case, 150 points can be awarded, entailing higher risk of high shock impedance and shock failure. The lateral view also allows to evaluate the PG placement in relation to the midline (Step 2 of the PRAETORIAN score). In (C), the S-ICD generator is in an optimal position, posterior to the midline; the score is therefore multiplied by 1. In (D), the entire generator is positioned anterior to the midline, and the score is multiplied by 2. The postero-anterior chest radiograph view allows to determine the amount of subgenerator fat (Step 3 of the PRATEORIAN score). As a reference, the generator width is used. In (E), less than 1 generator width of fat tissue is observed between the nearest point of the generator and the thoracic wall. The score is multiplied by 1. In (F), there is more than 1 generator width of fat tissue between the generator and the thoracic wall. The score is therefore multiplied by 1.5. DFT, defibrillation threshold; S-ICD, subcutaneous implantable cardioverter-defibrillator.
Based on the final score, three risk categories are defined:
Low risk of conversion failure: PRAETORIAN score of <90 points.
Intermediate risk of conversion failure: PRAETORIAN score between ≥90 and <150 points.
High risk of conversion failure: PRAETORIAN score of ≥150.
Chest radiographs were further analysed to assess (i) PG placement on PA chest radiograph view (Figure 3A): optimal if the PG is at the fifth or sixth intercostal space; inferior if the PG is below the sixth intercostal space; superior if the PG is above the fifth intercostal space, and (ii) electrode placement on PA chest radiograph view (Figure 3B): optimal if the proximal electrode is located in line with the xiphoid process; inferior if the proximal electrode is located below the xiphoid process; superior if the proximal electrode is located above the xiphoid process.13
Pulse generator (PG) and electrode position. (Row A). PG position on postero-anterior chest radiograph view. Optimal: the PG is at the fifth or sixth intercostal space (first column). Superior: the PG is above the fifth intercostal space (second column). Inferior: the PG is below the sixth intercostal space (third column). (Row B) Electrode position on postero-anterior chest radiograph view. Optimal: the proximal electrode is located in line with the xiphoid process (first column). Superior: the proximal electrode is located above the xiphoid process (second column). Inferior: the proximal electrode is located below the xiphoid process (third column).
Defibrillation testing
Defibrillation threshold testing was performed according to the local clinical practice and was not required for study enrolment. DFT tests were included in the analysis only if allowed to ascertain a defibrillation margin over the maximal ICD output of at least 15 J. More in detail, DFT testing efficacy was defined according to the following criteria:
any successful first shock ≤65 J was considered a successful test and was not repeated;
in case the first shock failed with standard polarity and was effective at the same energy with reverse polarity without the need for implant revision, the test was considered successful; and
in the case of a 65 J shock failure and further successful test either after implant revision or at >65 J, the test was considered failed.
Study aims
The primary aim of this study was to assess whether the implantation technique affects optimal S-ICD system position as defined by the PRAETORIAN score. The relationship between DFT, shock impedance and PRAETORIAN score was also explored.
Statistical analysis
Descriptive statistics are reported as mean ± standard deviation for normally distributed continuous variables, or medians with corresponding interquartile range (IQR) in the case of skewed distribution. Categorical variables are reported as percentages. Differences were compared by means of a t-test for Gaussian variables and Wilcoxon’s nonparametric test for non-Gaussian variables. The χ2 or Fisher’s exact test was used to compare proportions, as appropriate. Logistic regression analysis was used to determine the association between the PRAETORIAN score and the implantation technique. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated. All tests were two-sided and a P-value <0.05 was considered significant. All statistical analyses were performed by means of SPSS, version 21.
Results
Study population and subcutaneous implantable cardioverter-defibrillator implantation
During the observation period, 510 patients received an S-ICD at the 13 study centres. Of these, 213 had AP and lateral chest radiographs of adequate quality for PRAETORIAN score assessment and constituted the study cohort. Table 1 shows the baseline clinical variables of the study population. Patients were predominantly males (86%) and relatively young (48 ± 14 years). The most common reasons for implantation were Brugada syndrome (25.4%) and either ischaemic or idiopathic dilated cardiomyopathy with ejection fraction (EF) ≤35% (32.3%). When compared with the study cohort, patients excluded from the analysis because of unavailable or poor quality chest radiographs had less frequently Brugada syndrome (12%, P < 0.001) and low EF (48%; P < 0.001).
Clinical characteristics of the study population
| n = 213 | |
|---|---|
| Male gender, n (%) | 184 (86) |
| Age (years) | 48 ± 14 |
| Body mass index | 25.5 ± 3.7 |
| Heart disease, n (%) | |
| Brugada syndrome | 54 (25.4) |
| Ischaemic | 43 (20.2) |
| Dilated | 36 (16.9) |
| Hypertrophic | 33 (15.5) |
| ARVC | 19 (8.9) |
| Idiopathic VF | 12 (5.6) |
| Other | 16 (7.5) |
| LV ejection fraction (%) | 56 (35–65) |
| Chronic kidney disease, n (%) | 20 (9.4) |
| Diabetes, n (%) | 20 (9.4) |
| Beta blockers, n (%) | 111 (52.1) |
| Antiarrhythmic drugs, n (%) | 24 (11.3) |
| n = 213 | |
|---|---|
| Male gender, n (%) | 184 (86) |
| Age (years) | 48 ± 14 |
| Body mass index | 25.5 ± 3.7 |
| Heart disease, n (%) | |
| Brugada syndrome | 54 (25.4) |
| Ischaemic | 43 (20.2) |
| Dilated | 36 (16.9) |
| Hypertrophic | 33 (15.5) |
| ARVC | 19 (8.9) |
| Idiopathic VF | 12 (5.6) |
| Other | 16 (7.5) |
| LV ejection fraction (%) | 56 (35–65) |
| Chronic kidney disease, n (%) | 20 (9.4) |
| Diabetes, n (%) | 20 (9.4) |
| Beta blockers, n (%) | 111 (52.1) |
| Antiarrhythmic drugs, n (%) | 24 (11.3) |
ARVC, arrhythmogenic right ventricular cardiomyopathy; LV, left ventricular; VF, ventricular fibrillation.
Clinical characteristics of the study population
| n = 213 | |
|---|---|
| Male gender, n (%) | 184 (86) |
| Age (years) | 48 ± 14 |
| Body mass index | 25.5 ± 3.7 |
| Heart disease, n (%) | |
| Brugada syndrome | 54 (25.4) |
| Ischaemic | 43 (20.2) |
| Dilated | 36 (16.9) |
| Hypertrophic | 33 (15.5) |
| ARVC | 19 (8.9) |
| Idiopathic VF | 12 (5.6) |
| Other | 16 (7.5) |
| LV ejection fraction (%) | 56 (35–65) |
| Chronic kidney disease, n (%) | 20 (9.4) |
| Diabetes, n (%) | 20 (9.4) |
| Beta blockers, n (%) | 111 (52.1) |
| Antiarrhythmic drugs, n (%) | 24 (11.3) |
| n = 213 | |
|---|---|
| Male gender, n (%) | 184 (86) |
| Age (years) | 48 ± 14 |
| Body mass index | 25.5 ± 3.7 |
| Heart disease, n (%) | |
| Brugada syndrome | 54 (25.4) |
| Ischaemic | 43 (20.2) |
| Dilated | 36 (16.9) |
| Hypertrophic | 33 (15.5) |
| ARVC | 19 (8.9) |
| Idiopathic VF | 12 (5.6) |
| Other | 16 (7.5) |
| LV ejection fraction (%) | 56 (35–65) |
| Chronic kidney disease, n (%) | 20 (9.4) |
| Diabetes, n (%) | 20 (9.4) |
| Beta blockers, n (%) | 111 (52.1) |
| Antiarrhythmic drugs, n (%) | 24 (11.3) |
ARVC, arrhythmogenic right ventricular cardiomyopathy; LV, left ventricular; VF, ventricular fibrillation.
The S-ICD generator was positioned in an IM pocket in 174 patients (81.7%) and the two-incision approach was adopted in 199 (93.4%). A combination of the IM and two-incision technique was used in the majority of patients (n = 171; 80.3%). Out of 14 patients implanted with the three-incision technique, 3 (21.4%) had an IM pocket. On the contrary, the IM approach was adopted in the vast majority (171 out 199; 85.9%) of patients who underwent implantation with the two-incision technique.
The two- and three-incision techniques were equally distributed across quartiles of BMI (P = 0.79). On the contrary, a lower BMI was associated with higher probability of receiving IM PG implantation (HR 0.82; 95% CI 0.73–0.91; P < 0.001). Moreover, quartiles of patients’ enrolment in the registry predicted the use of the IM (HR 4.94; 95% CI 2.89–8.57; P < 0.001) and two-incision techniques (HR 4.05; 95% CI 1.75–9.35; P = 0.001).
PRAETORIAN score and implantation technique
According to the PRAETORIAN score, the risk of conversion failure was classified as low in 197 patients (92.5%), intermediate in 14 (6.6%), and high in 2 (0.9%). Patients with intermediate or high PRAETORIAN score had higher BMI than patients with low PRAETORIAN score. The other clinical characteristics were comparable between the two groups (Table 2). Figure 4 shows the impact of the implantation technique on the rate of optimal PG and lead position according to chest radiograph analysis. Patients undergoing the two-incision technique were more likely to have a low (<90) PRAETORIAN score than those implanted undergoing the three-incision technique (94.0% vs. 78.6%; P = 0.04) (Figure 4A). Using the two- or three-incision technique did not affect optimal infero-superior lead positioning (74.9% vs. 78.6%; P = 0.92).
Impact of the implantation technique on the rate of optimal PG and lead positioning according to chest radiograph analysis. (A) Patients undergoing the two-incision technique were more likely to have a low (<90) PRAETORIAN score than those undergoing the three-incision technique (P = 0.04). (B) Patients undergoing the IM technique were more likely to have a low (<90) PRAETORIAN score than those undergoing the subcutaneous approach (P = 0.001). PG, pulse generator.
Clinical and implant characteristics according to the PRAETORIAN score
| Low risk (n = 197) | High/ intermediate risk (n = 16) | P-value | |
|---|---|---|---|
| Male gender, n (%) | 169 (85.8) | 15 (93.8%) | 0.37 |
| Age (years) | 48 ± 14 | 49 ± 15 | 0.80 |
| Body mass index | 25.0 ± 3.4 | 30.8 ± 4.1 | <0.001 |
| LV EF <35%, n (%) | 64 (32.5) | 7 (43.8) | 0.35 |
| Implant | |||
| Shock impedance, Ohm (IQR) | 66 (59–77) | 96 (73–119) | 0.001 |
| Two-incision technique, n (%) | 187 (94.9) | 12 (75.0) | 0.002 |
| Intermuscular, n (%) | 167 (84.8) | 7 (43.8) | <0.001 |
| Low risk (n = 197) | High/ intermediate risk (n = 16) | P-value | |
|---|---|---|---|
| Male gender, n (%) | 169 (85.8) | 15 (93.8%) | 0.37 |
| Age (years) | 48 ± 14 | 49 ± 15 | 0.80 |
| Body mass index | 25.0 ± 3.4 | 30.8 ± 4.1 | <0.001 |
| LV EF <35%, n (%) | 64 (32.5) | 7 (43.8) | 0.35 |
| Implant | |||
| Shock impedance, Ohm (IQR) | 66 (59–77) | 96 (73–119) | 0.001 |
| Two-incision technique, n (%) | 187 (94.9) | 12 (75.0) | 0.002 |
| Intermuscular, n (%) | 167 (84.8) | 7 (43.8) | <0.001 |
EF, ejection fraction; IQR, interquartile range; LV, left ventricular; VF, ventricular fibrillation.
Clinical and implant characteristics according to the PRAETORIAN score
| Low risk (n = 197) | High/ intermediate risk (n = 16) | P-value | |
|---|---|---|---|
| Male gender, n (%) | 169 (85.8) | 15 (93.8%) | 0.37 |
| Age (years) | 48 ± 14 | 49 ± 15 | 0.80 |
| Body mass index | 25.0 ± 3.4 | 30.8 ± 4.1 | <0.001 |
| LV EF <35%, n (%) | 64 (32.5) | 7 (43.8) | 0.35 |
| Implant | |||
| Shock impedance, Ohm (IQR) | 66 (59–77) | 96 (73–119) | 0.001 |
| Two-incision technique, n (%) | 187 (94.9) | 12 (75.0) | 0.002 |
| Intermuscular, n (%) | 167 (84.8) | 7 (43.8) | <0.001 |
| Low risk (n = 197) | High/ intermediate risk (n = 16) | P-value | |
|---|---|---|---|
| Male gender, n (%) | 169 (85.8) | 15 (93.8%) | 0.37 |
| Age (years) | 48 ± 14 | 49 ± 15 | 0.80 |
| Body mass index | 25.0 ± 3.4 | 30.8 ± 4.1 | <0.001 |
| LV EF <35%, n (%) | 64 (32.5) | 7 (43.8) | 0.35 |
| Implant | |||
| Shock impedance, Ohm (IQR) | 66 (59–77) | 96 (73–119) | 0.001 |
| Two-incision technique, n (%) | 187 (94.9) | 12 (75.0) | 0.002 |
| Intermuscular, n (%) | 167 (84.8) | 7 (43.8) | <0.001 |
EF, ejection fraction; IQR, interquartile range; LV, left ventricular; VF, ventricular fibrillation.
Median PRAETORIAN score was 30.0 (IQR 30–45) for the IM and 45 (IQR 30–60) for the subcutaneous technique (P = 0.001). Patients undergoing the IM technique were more likely to have a low (<90) PRAETORIAN score than those undergoing the subcutaneous approach (96.0% vs. 79.5%; P = 0.001) (Figure 4B). The difference was mainly driven by the antero-posterior position of the S-ICD can, which was on or posterior to the midline in 96.0% of ‘IM patients’ and in 69.2% of ‘subcutaneous patients’ (P < 0.001). Patients undergoing the IM technique were more likely to have an optimal PG position on AP chest radiograph than those undergoing the subcutaneous approach; statistical significance was borderline (92.0% vs. 84.6%; P = 0.05).
The IM plus two-incision technique yielded a higher rate of PRAETORIAN score <90 than any other combination of implantation approaches (95.9% vs. 78.6%; P < 0.001). Body mass index was higher in patients with an intermediate/high-risk PRAETORIAN score than in those with a low score (30.8 ± 4.1 vs. 25.1 ± 3.4; P < 0.001) and was associated with an increased risk of being in the intermediate- or high-risk PRAETORIAN categories (HR 1.47; 95% CI 1.28–1.75; P < 0.001). After correction for BMI, the IM plus two-incision technique was associated with a low-risk PRAETORIAN score (HR 3.76; 95% CI 1.01–14.02; P = 0.04). Median PRAETORIAN score was comparable among quartiles of enrolment in the registry (30 for each group; P = 0.48).
DFT, shock impedance, and PRAETORIAN score
Twenty-four (11%) patients did not undergo DFT testing and 2 (0.9%) were tested at ≤40 J. Patients who did not underwent DFT testing did not differ from those who were tested (males: 83 vs. 87%, P = 0.75; mean age: 51 ± 16 vs. 48 ± 14 years, P = 0.31; BMI: 25.7 ± 2.3 vs. 25.5 ± 3.8; P = 0.81) and had similar PRAETORIAN score [30 (IQR 30–41) vs. 30 (IQR 30–60); P = 0.41]. Of the 187 patients tested at 65 J, 179 (95.7%) were successfully converted to sinus rhythm (nine with reverse polarity). In seven patients in whom the 65 J shock failed, a second shock between 70 and 80 J was effective. One patient with a first failed shock underwent lead repositioning and was successfully defibrillated at 65 J after a second DFT test. In the two patients tested at ≤40 J, the first S-ICD shock was ineffective. A second (80 J) shock converted ventricular fibrillation (VF) to sinus rhythm. Overall, median shock impedance was 68 Ohm (IQR 48–88).
There were no clinical or chest radiograph predictors of shock failure at 65 J. On univariate analysis, the PRAETORIAN score did not predict shock failure at 65 J, neither as a continuous (HR 1.00; 95% CI 0.98–1.01; P = 0.92), nor as a categorical variable (low vs. intermediate or high risk, HR 0.50; 95% CI 0.05–4.43; P = 0.53). Out of eight shock failures, one occurred in the first quartile of study enrolment, three in the second, two in the third, and two in the fourth (P = 0.73).
Shock impedance was lower in PRAETORIAN low-risk patients than in intermediate- or high-risk categories (66 Ohm, IQR 59–77 vs. 96 Ohm, IQR 73–119; P = 0.001) and was associated with an intermediate-/high-risk PRAETORIAN score after correction for BMI (HR 1.05; 95% CI 1.01–1.09; P = 0.003). The two-incision plus IM technique yielded lower shock impedance than any other combination of implantation approaches (67 Ohm, IQR 59–77 vs. 72 Ohm, IQR 65–91; P = 0.01). Median shock impedance (72, 70, 68, and 62 Ohm; P = 0.01) decreased significantly from the first to the last quartile of study enrolment.
Discussion
PRAETORIAN score and implantation technique
In this study, we evaluated whether the S-ICD implantation technique affected the proper system position on chest radiograph. We found that patients undergoing a combination of the IM and two-incision technique had a higher probability of being at low risk of shock failure according to the PRAETORIAN score. Moreover, we confirmed that a lower PRAETORIAN score was associated with lower shock impedance.
The PRAETORIAN score was developed as a practical tool for predicting shock efficacy after S-ICD implantation6 and provides feedback on which element of the implantation procedure contributes to the risk of conversion failure. Indeed, it takes into account three major determinants of S-ICD DFT: (i) adipose tissue between the coil and the sternum, (ii) anterior PG position along the AP axis, causing the current to shunt through the anterior chest wall, and (iii) adipose tissue between the generator and the thorax. Amin at al. also reported on the critical impact of S-ICD position on shock efficacy, showing that inferior system placement and superficial coil implantation were associated with shock failure.13 Of note, although shock impedance is correlated with conversion success, high current flow through the heart requires that sufficient cardiac mass be included between the coil and the can, underlining the need for an optimal anatomic approach to implantation.
The S-ICD implantation technique has changed significantly over time, with the introduction of the two-incision technique and IM generator implantation. Both approaches have been shown to be safe and effective, and to offer better cosmetic outcomes and shorter procedural times.8,9,14,15 In our study cohort, the vast majority of patients had a PRAETORIAN score <90. This finding is in line with previous observations.6 The IM plus two-incision technique resulted in a lower PRAETORIAN score. Indeed, with this approach, the coil appears to be implanted more deeply via the dedicated sheath introducer and the PG positioned in contact with the muscular fascia, thus preventing the interposition of fat tissue. Moreover, IM implantation requires that the PG be positioned under the latissimus dorsi muscle, which runs posteriorly to the mid-axillary line, thus preventing anterior malpositioning of the S-ICD can. All of these factors affect the individual analytics that compose the PRAETORIAN score. Of note, neither the two-incision nor the IM technique caused the placement of the coil or the PG to be too inferior or superior, two determinants of shock failure.13
Predictors of shock efficacy
Shock efficacy was high and comparable to that observed in previous studies.16–18 Two patients were tested at ≤40 J and, after failure, at 80 J. This prevented us from assessing DFT between 40 and 65 J. As only eight patients had shock failure at 65 J, we were unable to reliably assess the predictive value of clinical, surgical or chest radiograph variables on shock efficacy. Indeed, the probabilistic mechanism to DFT testing requires that sufficient test outcomes are collected to overcome the stochastic nature of combining multiple variables that affect DFT (e.g. tissue ischaemia during VF, autonomic balance, electrolyte abnormalities): if defibrillation is successful at a given energy level, it may not be effective on a second attempt.19
High-voltage shock impedance
Increased shock impedance has been observed in association with failed conversion testing, and is regarded as a predictor of shock failure.6 Indeed, as stated by the Ohm’s law, for a given energy level, the current delivered to the myocardium is inversely proportional to the transthoracic resistance (or impedance in the case of alternating current). Thus, lowering the shock impedance helps to ensure optimal current delivery. Proper contact between the coil and the sternum and the PG and the thoracic wall minimizes fat interposition and, therefore, lowers the impedance, thus granting effective defibrillation energy. Accordingly, Amin et al.13 have reported that high shock impedance is associated with inferiorly positioned coil and PG, inadequate coil depth and lower rate of defibrillator success. Notwithstanding this, it should be emphasized that coil or PG malposition (e.g. a coil excessively leftward to the sternum or too anterior PG) may result in normal shock impedance but energy shunt through the chest wall, entailing low current through the heart. Thus, avoiding both fat interposition at the coil/sternum or PG/thorax interface and energy shunt through the chest wall are necessary to warrant low DFT. In our study, shock impedance was lower in PRAETORIAN low-risk patients than in intermediate- or high-risk categories, indicating a higher probability of effective defibrillation. This is in agreement with previous observations.6 In line with this, the two-incision plus IM technique yielded lower shock impedance than any other combination of coil and generator implantation approaches. Moreover, patients undergoing the IM technique were also more likely to have the PG implanted on or posterior to the midline as compared to subcutaneous patients. The combination of low shock impedance and posterior PG implantation is a crucial prerequisite for lowering the DFT.
Study limitations
This study has several limitations. First, this was a retrospective non-randomized study. Therefore, conversion testing and chest X-ray protocols varied among implanting centres. Moreover, patients body habitus may have affected the selection of the implanting technique. Nonetheless, patients were consecutively enrolled, and the data prospectively collected. High-voltage impedances were not available in all patients. Second, the PRAETORIAN score was intermediate-to-high in a small proportion of patients. This might have affected the statistical power of our analysis. Third, differences in PRAETORIAN score among groups may be due to several factors. Particularly, the progresses in general approach to S-ICD implantation over time is difficult to separate from the changes in implant technique. Ascribing the improvements to the latter ignores the effects of the former. Fourth, the number of obese patients was too small to assess whether the three-incision technique yields lower PRAETORIAN score or shock impedance in this specific subgroup. Indeed, in obese patients, the three-incision technique may ensure deeper tunnelling and anchoring of the lead. Finally, this cohort had fewer comorbidities and less structural heart disease than typical ICD cohorts.
Conclusions
In this cohort of S-ICD recipients, combining the two-incision technique and IM generator implantation yielded the lowest PRAETORIAN scores and shock impedance values, indicating optimal defibrillation system position and a high probability of effective defibrillation.
Conflict of interest: Pietro Francia received speaker’s fees and institutional grants from Boston Scientific and research grants from Abbott. Maurizio Landolina received fees as speaker and advisory board member from Boston Scientific, LivaNova (now Microport), and Medtronic. Mariolina Lovecchio and Sergio Valsecchi are employees of Boston Scientific. All other authors have declared no conflict of interests.
Data availability
The data can be shared on reasonable request to the corresponding author.
