Risk stratification of patients undergoing transvenous lead extraction with the ELECTRa Registry Outcome Score (EROS): an ESC EHRA EORP European lead extraction ConTRolled ELECTRa registry analysis

Abstract

Aims

Transvenous lead extraction is associated with a significant risk of complications and identifying patients at highest risk pre-procedurally will enable interventions to be planned accordingly. We developed the ELECTRa Registry Outcome Score (EROS) and applied it to the ELECTRa registry to determine if it could appropriately risk-stratify patients.

Methods and results

EROS was devised to risk-stratify patients into low risk (EROS 1), intermediate risk (EROS 2), and high risk (EROS 3). This was applied to the ESC EORP European Lead Extraction ConTRolled ELECTRa registry; 57.5% EROS 1, 31.8% EROS 2, and 10.7% EROS 3. Patients with EROS 3 or 2 were significantly more likely to require powered sheaths and a femoral approach to complete procedures. Patients with EROS 3 were more likely to suffer procedure-related major complications including deaths (5.1 vs. 1.3%; P < 0.0001), both intra-procedural (3.5 vs. 0.8%; P = 0.0001) and post-procedural (1.6 vs. 0.5%; P = 0.0192). They were more likely to suffer post-procedural deaths (0.8 vs. 0.2%; P 0.0449), cardiac avulsion or tear (3.8 vs. 0.5%; P < 0.0001), and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (4.6 vs. 1.0%; P < 0.0001). EROS 3 was associated with procedure-related major complications including deaths [odds ratio (OR) 3.333, 95% confidence interval (CI) 1.879–5.914; P < 0.0001] and all-cause in-hospital major complications including deaths (OR 2.339, 95% CI 1.439–3.803; P = 0.0006).

Conclusion

EROS successfully identified patients who were at increased risk of significant procedural complications that require urgent surgical intervention.

ELECTRa registry, ELECTRa Registry Outcome Score, Risk stratification, Transvenous lead extraction

What’s new?

• Identifying patients at an increased risk of complications following transvenous lead extraction helps plan procedures and improves outcomes.

• An ELECTRa Registry Outcome Score (EROS) was devised to risk stratify patients into low risk (1), intermediate risk (2), and high risk (3).

• Patients with EROS 3 were significantly more likely to have procedures requiring powered sheaths, femoral approach and suffer procedure-related major complications including deaths, cardiac avulsion or tear and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair.

• EROS 3 was associated with procedure-related major complications including deaths and all-cause in-hospital major complications including deaths.

• The EROS risk-score can appropriately identify patients at increased risk of significant procedural and in-hospital complications.

Introduction

Transvenous lead extraction (TLE) is important in the management of infected, redundant or malfunctioning pacemaker and defibrillator leads.^1–3 Major complications require prompt surgical intervention with a thoracotomy or sternotomy and guidelines recommend this is undertaken within 5–10 min.² High-risk procedures carried out in the catheterization laboratory compared with the operating room are associated with an increased risk of death.⁴ Therefore, appropriate pre-procedural risk stratification is essential to ensure high-risk procedures are undertaken in an appropriate setting with relevant personnel present.

The ESC EORP ELECTRa registry provided contemporary outcomes of TLE in European centres⁵ reporting a 1.7% rate of procedure-related major complications including deaths and 0.5% rate of procedure-related deaths. A sub-analysis of the ELECTRa registry examined outcomes according to the centre volume and extraction location and demonstrated that procedures performed by primary operator cardiologists in low-volume centres were associated with increased procedure-related major complications including deaths [odds ratio (OR) 1.858, 95% confidence interval (CI) 1.007–3.427; P = 0.0475].⁶ Indeed, rates of procedure-related major complications including deaths in high-risk areas, defined as the hybrid theatre or operating room, were similar to those of the catheterization laboratory (1.7 vs. 1.6%; P = 0.9297), suggesting further risk stratification is required.

Risk stratification scores have been previously studied.^1,2,7 Kancharla et al.⁸ allocated high-risk patients to the operating room and intermediate-risk patients to the device laboratory. They prospectively enrolled 187 patients and demonstrated there was a significantly higher rate of major complications in the operating room compared with the device laboratory (6.9 vs. 0.0%; P = 0.007). The main strengths of this study was its prospective design, validation in procedures involving simple and complex tools and ability to identify patients at an increased risk of major complications. However, this was only a single-centre study and further validation is required in a larger cohort, to ensure this score is reproducible and truly high-risk patients are appropriately identified.

We developed a risk-stratification score based on previous studies and retrospectively applied this to the ELECTRa registry, to determine whether it could identify high-risk cases that subsequently developed significant major complications.

Methods

The design and results of the ELECTRa study has been previously published.⁵ The executive committee in collaboration with the EURObservational Research Programme (EORP) provided the study design, protocol and scientific leadership of the registry under the responsibility of the European Heart Rhythm Association Scientific Initiative Committee. Overall, 3510 patients underwent TLE in 73 European centres and lead dwell time was available in 3485 patients.⁵ High-volume (HiV) centres were defined as ≥30 TLE procedures/year and low-volume (LoV) centres as <30 TLE procedures/year. In the current analysis, we developed a risk score based on a previous prospective study,⁸ and our experience to pre-procedurally stratify patients undergoing TLE. Patients were assigned an ELECTRa Registry Outcome Score (EROS) 1, 2, or 3 depending on whether their predicted risk of major complications was low, intermediate, or high (Table 1). The score was based on patient characteristics, comorbidities, lead characteristics, and lead dwell time. We applied this score to the ELECTRa registry and compared baseline patient demographics, lead and procedural outcomes. In this analysis, we compared characteristics and outcomes of EROS 3 with both EROS 1 and 2 combined and also EROS 1 with EROS 2, since this would help identify whether the risk score could distinguish between high- and intermediate-risk patients. Overall, 2004/3485 (57.5%) patients were EROS 1, 1109/3485 (31.8%) EROS 2, and 372/3485 (10.7%) EROS 3. The operating room or hybrid theatre was both considered a high-risk setting since this environment can facilitate urgent surgical intervention and the catheterization laboratory a low-risk setting.

Table 1

Risk score

EROS 1	EROS 2	EROS 3
Pacemaker lead ≤15 years from implant	Pacemaker lead ≤15 years or ICD lead ≤10 years from implant and either:	Pacemaker lead > 15 years from implant
ICD lead ≤10 years from implant	Congenital heart disease Initial implant when the patient was <15 years old Chronic kidney disease and serum creatinine >2 mg/dL Infectious indication for extraction and any one of the following: White cell count > 12 × 109/L Positive blood culturea Vegetation on transoesophageal echocardiogram Chronic heart failure and New York Heart Association functional class IV	ICD lead > 10 years from implant

EROS, ELECTRa Registry Outcome Score; ICD, implantable cardioverter-defibrillator.

a

Organisms include any gram negative or positive bacteria, fungal, or polymicrobial.

Declaration of Helsinki

This study complies with the Declaration of Helsinki and locally appointed ethics committees have approved the research.

Statistical analysis

The results are presented as mean ± standard deviation for normally distributed variables and as median (interquartile range) for non-normally distributed variables. A X² was used for among-group comparisons or Fisher’s exact test if the expected cell count was less than five. Continuous non-normally distributed data were compared using Mann–Whitney. Univariable and multivariable binary logistic regression was performed to determine predictors of procedure and all-cause in-hospital major complications including deaths. Variables statistically significant, with a P-value <0.05⁹ at univariable analysis were used for multivariable analysis. A two-sided P-value <0.05 was statistically significant. Analyses were performed using Stata (StataCorp 2015, Stata Statistical Software: Release 14; StataCorp LP, USA), SPSS (IBM Switzerland, Version 26, Switzerland), and Prism (GraphPad Software Inc., Version 8, CA, USA).

Results

Baseline patient demographics

Baseline demographics according to EROS are provided in Table 2. Patients with EROS 3 compared with EROS 1 and 2 combined, were more likely to be female (32.5 vs. 27.1%; P = 0.0264), have a lower body mass index (BMI) [25.7 (23.5–28.0) vs. 26.2 (23.6–29.4) kg/m²; P = 0.0161], infectious indication (61.9 vs. 52.2%; P = 0.0004), and longer lead dwell time [16 (12–19) vs. 5 (2–7) years; P < 0.0001] but less likely to have chronic heart failure (32.6 vs. 46.2%; P < 0.0001). Patients with EROS 2 compared with 1 were more likely to be older (65.2 ± 17.9 vs. 64.7 ± 14.2 years; P = 0.0001), have a lower BMI [25.8 (23.3–28.7) vs. 26.4 (23.8–29.4) kg/m²; P = 0.0001] and have an infectious indication (86.2 vs. 33.2%; P < 0.0001) but less likely to be female (24.8 vs. 28.3%; P = 0.0330).

Table 2

Baseline patient demographics according to the patients ELECTRa Registry Outcome Score

Variable	Total	EROS 1	EROS 2	EROS 3	P-valuea	P-valueb
Characteristics
Age (years), ±SD	64.90 ± 15.63	64.74 ± 14.23	65.16 ± 17.92	64.97 ± 15.54	0.9594	0.0001
Female, n/N (%)	964/3485 (27.66)	568/2004 (28.34)	275/1109 (24.80)	121/372 (32.53)	0.0264	0.0330
Body mass index (kg/m2), median (IQR)	26.10 (23.55–29.30)	26.40 (23.80–29.40)	25.75 (23.33–28.70)	25.70 (23.50–28.00)	0.0161	0.0001
Left ventricular ejection fraction (%), median (IQR)	50.00 (33.00–60.00)	46.00 (32.00–60.00)	49.00 (32.00–60.00)	52.00 (40.00–60.00)	<0.0001	0.8152
Co-morbidities, n/N (%)
Coronary artery disease	1365/3459 (39.46)	792/1991 (39.78)	459/1098 (41.80)	114/370 (30.81)	0.0003	0.2726
Hypertension	1876/3455 (54.30)	1092/1986 (54.98)	600/1101 (54.50)	184/368 (50.00)	0.0799	0.7937
Diabetes mellitus	777/3464 (22.43)	395/1991 (19.84)	325/1103 (29.47)	57/370 (15.41)	0.0006	<0.0001
Chronic heart failure	1549/3464 (44.72)	930/1993 (46.66)	498/1100 (45.27)	121/371 (32.61)	<0.0001	0.4577
Dilated cardiomyopathy	911/3467 (26.28)	554/1996 (27.76)	293/1101 (26.61)	64/370 (17.30)	<0.0001	0.4944
Chronic obstructive pulmonary disease	295/3459 (8.53)	147/1991 (7.38)	125/1100 (11.36)	23/368 (6.25)	0.0978	0.0002
Indication, n/N (%)
Infection	1848/3476 (53.16)	663/1997 (33.20)	956/1109 (86.20)	229/370 (61.89)	0.0004	<0.0001
Local infection	1160/3476 (33.37)	617/1997 (30.90)	398/1109 (35.89)	145/370 (39.19)	0.0121	0.0045
Systemic infection	673/3476 (19.36)	33/1997 (1.65)	558/1109 (50.32)	82/370 (22.16)	0.1492	<0.0001
No infection	1637/3485 (46.97)	1341/2004 (66.92)	153/1109 (13.80)	143/372 (38.44)	0.0005	<0.0001
Lead dwell time (years), median (IQR)	5.33 (3.00–8.00)	5.00 (2.00–7.00)	5.00 (2.00–7.33)	16.00 (12.00–19.00)	<0.0001	0.4356

EROS, ELECTRa Registry Outcome Score; IQR: interquartile range; SD: standard deviation.

a

EROS 3 vs. both EROS 1 and 2 combined.

b

EROS 2 vs. EROS 1.

Procedural characteristics and clinical success

Procedural and lead outcomes are provided in Table 3. Patients with EROS 3 compared with EROS 1 and 2 combined, were more likely to require powered sheaths (44.9 vs. 29.2%; P < 0.0001), including both laser sheaths (30.7 vs. 20.3%; P < 0.0001) and Evolution^® mechanical dilator sheaths (14.8 vs. 8.9%; P = 0.0003). They were more likely to require a femoral approach (8.3 vs. 3.8%; P = 0.0001) but less likely to achieve clinical success (89.0 vs. 97.7%; P < 0.0001). They were more likely to require a prolonged procedure time [114 (76–160) vs. 80 (55–120) min; P < 0.0001], extraction time [35 (15–60) vs. 15 (5–34) min; P < 0.0001], and hospital stay related to TLE [4 (2–9) vs. 3 (2–7) days; P = 0.0002].

Table 3

Procedural and lead outcomes with different operators and extraction environments according to the patients ELECTRa Registry Outcome Score

Variable	Total	EROS 1	EROS 2	EROS 3	P-valuea	P-valueb
Sheaths, n/N (%)
Powered sheaths (any)	1077/3485 (30.90)	537/2004 (26.80)	373/1109 (33.63)	167/372 (44.89)	<0.0001	0.0001
Laser sheaths	745/3485 (21.38)	348/2004 (17.37)	283/1109 (25.52)	114/372 (30.65)	<0.0001	<0.0001
Evolution® mechanical dilator sheaths	332/3485 (9.53)	189/2004 (9.43)	88/1109 (7.94)	55/372 (14.78)	0.0003	0.1603
Electrosurgical dissection sheaths	5/3485 (0.14)	1/2004 (0.05)	3/1109 (0.27)	1/372 (0.27)	0.4315	0.1324
Femoral approach for at least one lead, n/N (%)	150/3485 (4.30)	65/2004 (3.24)	54/1109 (4.87)	31/372 (8.33)	0.0001	0.0235
Clinical success, n/N (%)	3371/3485 (96.73)	1961/2004 (97.85)	1079/1109 (97.29)	331/372 (88.98)	<0.0001	0.3233
Duration per patients, median (IQR)
Procedure time (min)	82.00 (57.00–120.00)	80.00 (58.25–120.00)	75.00 (50.00–120.00)	114.00 (76.00–160.00)	<0.0001	0.0105
Extraction time (min)	19.00 (5.00–40.00)	15.00 (5.00–30.00)	20.00 (5.00–40.00)	35.00 (15.00–60.00)	<0.0001	0.0002
Duration of hospital stay related to lead extraction (days), ±SD	3.00 (2.00–7.00)	3.00 (2.00–5.00)	5.00 (2.00–13.00)	4.00 (2.00–9.00)	0.0002	<0.0001
High-risk environment, n/N (%)
Overall	2148/3485 (61.64)	1285/2004 (64.12)	613/1109 (55.28)	250/372 (67.20)	0.0194	<0.0001
High-volume centre	1872/2859 (65.48)	1161/1693 (68.58)	490/852 (57.51)	221/314 (70.38)	0.0527	<0.0001
Low-volume centre	276/626 (44.09)	124/311 (39.87)	123/257 (47.86)	29/58 (50.00)	0.3412	0.0559
Low-risk environment, n/N (%)
Overall	1337/3485 (38.36)	719/2004 (35.88)	496/1109 (44.72)	122/372 (32.80)	0.0194	<0.0001
High-volume centre	987/2859 (34.52)	532/1693 (31.42)	362/852 (42.49)	93/314 (29.62)	0.0527	<0.0001
Low-volume centre	350/626 (55.91)	187/311 (60.13)	134/257 (52.14)	29/58 (50.00)	0.3412	0.0559
Primary operator cardiologist, n/N (%)	3225/3485 (92.54)	1874/2004 (93.51)	1010/1109 (91.07)	341/372 (91.67)	0.4979	0.0125
Primary operator surgeon, n/N (%)	257/3485 (7.37)	128/2004 (6.39)	98/1109 (8.84)	31/372 (8.33)	0.4540	0.0117
Cardiologist and surgeons present, n/N (%)	546/3483 (15.68)	315/2003 (15.73)	170/1108 (15.34)	61/372 (16.40)	0.6854	0.7777
General anaesthesia at the start of the procedure, n/N (%)	1339/3485 (38.42)	680/2004 (33.93)	448/1109 (40.40)	211/372 (56.72)	<0.0001	0.0003
Arterial line placed at the start of the procedure, n/N (%)	2029/3485 (58.22)	1018/2004 (50.80)	741/1109 (66.82)	270/372 (72.58)	<0.0001	<0.0001

EROS, ELECTRa Registry Outcome Score; IQR: interquartile range; SD: standard deviation.

a

EROS 3 vs. both EROS 1 and 2 combined.

b

EROS 2 vs. EROS 1.

Patients with EROS 2 compared with 1 were more likely to require the use of powered sheaths (33.6 vs. 26.8%; P = 0.0001), particularly laser sheaths (25.5 vs. 17.4%; P < 0.0001) and require a femoral approach (4.9 vs. 3.2%; P = 0.0235). They were more likely to need a prolonged extraction time [20 (5–40) vs. 15 (5–30) min; P = 0.0002] and hospital stay related to TLE [5 (2–13] vs. 3 (2–5) days; P < 0.0001]. They were matched in terms of clinical success (97.3 vs. 97.9%; P = 0.3233).

Extraction environment and operators

Procedural outcomes according to different extraction environments and operators are provided in Table 3. Patients with EROS 3 compared with EROS 1 and 2 combined, were more likely to have their procedure in a high-risk environment (67.2 vs. 61.0%; P = 0.0194) and at the start of procedures require general anaesthesia (56.7% vs. 36.2%; P < 0.0001) and an arterial line (72.6 vs. 56.5%; P < 0.0001). EROS 2 compared with 1 were more likely to be undertaken in a low-risk environment (44.7 vs. 35.9%; P < 0.0001), be performed by a primary operator surgeon (8.8 vs. 6.4%; P = 0.0117) and were less likely to be performed by a primary operator cardiologist (91.1 vs. 93.5%; P = 0.0125). At the start of procedures, they were also more likely to require general anaesthesia (40.4 vs. 33.9%; P = 0.0003) and an arterial line (66.8 vs. 50.8%; P < 0.0001).

Procedural complications

Procedural complications are provided in Table 4. Patients with EROS 3 compared with EROS 1 and 2 combined, were more likely to suffer procedure-related major complications including deaths (5.1 vs. 1.3%; P < 0.0001), both intra-procedural (3.5 vs. 0.8%; P = 0.0001) and post-procedural (1.6 vs. 0.5%; P = 0.0192). In particular, they were more likely to suffer post-procedural deaths (0.8 vs. 0.2%; P = 0.0449), cardiac avulsion or tear (3.8 vs. 0.5%; P < 0.0001) and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (4.6 vs. 1.0%; P < 0.0001). Additionally, they were more likely to suffer all-cause in-hospital major complications including deaths (6.5 vs. 2.3%; P < 0.0001) and minor complications (8.3 vs. 4.6%; P = 0.0015). At multivariable binary logistic regression, EROS 3 was associated with procedure-related major complications including deaths (OR 3.333, 95% CI 1.879–5.914; P < 0.0001), intra-procedural-related major complications including deaths (OR 4.041, 95% CI 2.011–8.119; P = 0.0001), cardiac avulsion or tear (OR 7.111, 95% CI 3.382–14.949; P < 0.0001), and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (OR 3.860, 95% CI 2.095–7.113; P < 0.0001) (Figure 1 and 2). EROS 3 was also associated with all-cause in-hospital major complications including deaths (OR 2.339, 95% CI 1.439–3.803; P = 0.0006) and minor complications (OR 1.588, 95% CI 1.049–2.404; P = 0.0288). In addition, EROS 3 patients who underwent TLE in a high-risk vs. low-risk environment were matched in terms of procedure-related major complications including deaths (5.6 vs. 4.1%; P = 0.5368), both intra-procedural (4.0 vs. 2.5%; P = 0.5584) and post-procedural (2.0 vs. 1.6%; P = 1.0000). In patients with EROS 3 compared with EROS 1 and 2 who underwent TLE in a HiV centre and high-risk environment, there was a significantly higher rate of major complications including deaths (5.4 vs. 1.0%; P < 0.0001), including both intra-procedural (3.6 vs. 0.6%; P = 0.0005) and post-procedural (1.8 vs. 0.4%; P = 0.0223) (Supplementary material online, Appendix S2).

Figure 1

Predictors of procedure-related major complications including deaths and intra-procedure-related major complications including deaths. CI, confidence intervals; EROS, ELECTRa Registry Outcome Score. ^aAdjusted for all variables shown.

Figure 2

Predictors of procedure-related major complications. CI, confidence intervals; EROS, ELECTRA Registry Outcome Score. ^aAdjusted for all variables shown.

Table 4

Procedural outcomes according to the patients ELECTRa Registry Outcome Score

Variable	Total	EROS 1	EROS 2	EROS 3	P-valuea	P-valueb
Procedure-related major complications including deaths, n/N (%)	58/3485 (1.66)	24/2004 (1.20)	15/1109 (1.35)	19/372 (5.11)	<0.0001	0.7097
Intra-procedural	37/3485 (1.06)	16/2004 (0.80)	8/1109 (0.72)	13/372 (3.49)	0.0001	0.8140
Post-procedural	21/3485 (0.60)	8/2004 (0.40)	7/1109 (0.63)	6/372 (1.61)	0.0192	0.3707
Details of procedure-related major complications including deaths, n/N (%)
Procedural-related deaths	17/3485 (0.49)	6/2004 (0.30)	7/1109 (0.63)	4/372 (1.08)	0.0993	0.2438
Intra-procedural	9/3485 (0.26)	5/2004 (0.25)	3/1109 (0.27)	1/372 (0.27)	1.0000	1.0000
Post-procedural	8/3485 (0.23)	1/2004 (0.05)	4/1109 (0.36)	3/372 (0.81)	0.0449	0.0574
Cardiac avulsion or tear	30/3485 (0.86)	9/2004 (0.45)	7/1109 (0.63)	14/372 (3.76)	<0.0001	0.4962
Vascular avulsion or tear	20/3485 (0.57)	11/2004 (0.55)	4/1109 (0.36)	5/372 (1.34)	0.0543	0.4677
Cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair	49/3485 (1.41)	20/2004 (1.00)	12/1109 (1.08)	17/372 (4.57)	<0.0001	0.8238
Heart failure	1/3485 (0.03)	0/2004 (0.00)	0/1109 (0.00)	1/372 (0.27)	0.1067	–
Sepsis	1/3485 (0.03)	0/2004 (0.00)	0/1109 (0.00)	1/372 (0.27)	0.1067	–
Respiratory arrest	2/3485 (0.06)	2/2004 (0.10)	0/1109 (0.00)	0/372 (0.00)	1.0000	0.5412
Multiorgan failure	1/3485 (0.03)	0/2004 (0.00)	1/1109 (0.09)	0/372 (0.00)	1.0000	0.3562
Cerebrovascular accident	2/3485 (0.06)	1/2004 (0.05)	1/1109 (0.09)	0/372 (0.00)	1.0000	1.0000
Arrhythmias	1/3485 (0.03)	0/2004 (0.00)	1/1109 (0.09)	0/372 (0.00)	1.0000	0.3562
Acute superior vena cava syndrome	1/3485 (0.03)	0/2004 (0.00)	1/1109 (0.09)	0/372 (0.00)	1.0000	0.3562
Anaesthesia-related complications	2/3485 (0.06)	2/2004 (0.10)	0/1109 (0.00)	0/372 (0.00)	1.0000	0.5412
Acute abdominal occlusion	1/3485 (0.03)	0/2004 (0.00)	0/1109 (0.00)	1/372 (0.27)	0.1067	–
Disseminate intravascular coagulation	1/3485 (0.03)	0/2004 (0.00)	1/1109 (0.09)	0/372 (0.00)	1.0000	0.3562
All-cause in-hospital major complications including deaths, n/N (%)	95/3485 (2.73)	29/2004 (1.45)	42/1109 (3.79)	24/372 (6.45)	<0.0001	<0.0001
All-cause deaths	50/3485 (1.43)	10/2004 (0.50)	32/1109 (2.89)	8/372 (2.15)	0.2193	<0.0001
Details of non-procedure-related major complications including death, n/N (%)
Heart failure	15/3485 (0.43)	2/2004 (0.10)	12/1109 (1.08)	1/372 (0.27)	1.0000	0.0002
Sepsis	11/3485 (0.32)	1/2004 (0.05)	7/1109 (0.63)	3/372 (0.81)	0.1042	0.0040
Respiratory arrest	5/3485 (0.14)	1/2004 (0.05)	3/1109 (0.27)	1/372 (0.27)	0.4315	0.1324
Multiorgan failure	4/3485 (0.11)	0/2004 (0.00)	4/1109 (0.36)	0/372 (0.00)	1.0000	0.0161
Cerebrovascular accident	2/3485 (0.06)	1/2004 (0.05)	1/1109 (0.09)	0/372 (0.00)	1.0000	1.0000
Arrhythmias	2/3485 (0.06)	1/2004 (0.05)	1/1109 (0.09)	0/372 (0.00)	1.0000	1.0000
Deaths	33/3485 (0.95)	4/2004 (0.20)	25/1109 (2.25)	4/372 (1.08)	0.7747	<0.0001
Minor complications	173/3485 (4.96)	82/2004 (4.09)	60/1109 (5.41)	31/372 (8.33)	0.0015	0.0913

EROS, ELECTRa Registry Outcome Score.

a

EROS 3 vs. both EROS 1 and 2 combined.

b

EROS 2 vs. EROS 1.

Patients with EROS 2 compared with 1 were matched in terms of procedure-related major complications including deaths (1.4 vs. 1.2%; P = 0.7097), including cardiac avulsion or tear (0.6 vs. 0.5%; P = 0.4962) and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (1.1 vs. 1.0%; P = 0.8238). However, they were more likely to suffer all-cause in-hospital major complications including deaths (3.8 vs. 1.5%; P < 0.0001) and all-cause deaths (2.9 vs. 0.5%; P < 0.0001).

Discussion

Risk stratification for TLE is essential to ensure high-risk patients have their procedure performed in an environment that can facilitate urgent surgical intervention. We developed a risk score (EROS) to identify TLE patients at increased risk of complications and applied this to the ELECTRa registry. The predominant benefit of this score over other TLE risk-scores is its design and validation in the ELECTRa registry, the largest cohort of patients providing contemporary outcomes of TLE procedures in multiple European centres with varied experience. We demonstrated that:

• EROS 3 were more likely to require a femoral approach, powered sheaths, have a prolonged procedure time, extraction time, length of hospital admission and were less likely to achieve clinical success.

• EROS 3 were more likely to experience procedure-related major complications including deaths, including both intra- and post-procedure. EROS 3 were independently associated with procedure-related major complications including deaths (OR 3.333, 95% CI 1.879–5.914; P < 0.0001), cardiac avulsion or tear (OR 7.111, 95% CI 3.382–14.949; P < 0.0001), and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (OR 3.860, 95% CI 2.095–7.113; P < 0.0001).

• EROS 2 compared with 1 were more likely to require powered sheaths, femoral approach to complete procedures, have a prolonged extraction time and length of hospital admission. They had similar procedure-related major complications including deaths but EROS 2 had a significantly higher proportion of all-cause in-hospital major complications including deaths.

Identifying patients at an increased risk of complications

Appropriately identifying patients at an increased risk of major complications pre-procedurally allows appropriate allocation of resources to ensure TLE is performed in the most appropriate environment with relevant personnel present. Several studies have investigated different baseline characteristics that predict a higher risk of complications following TLE including advanced age, BMI <25 kg/m², severe left ventricular systolic impairment, chronic kidney disease, infection, large vegetations, congenital heart disease, lead type, and a prolonged lead dwell time.^4,5,10 The EROS risk score adds to this body of evidence for risk stratification and demonstrated that patients with EROS 3 compared with EROS 1 and 2 combined, were significantly more likely to be female, have a lower BMI, infectious indication for TLE and have a prolonged lead dwell time. The use of powered sheaths and a femoral approach to complete procedures are associated with an increased risk of procedure-related major complications.^5,11 Identifying patients that are likely to require these interventions is essential and patients who were EROS 3 or 2, were significantly more likely to require these interventions.

Risk scores

The development of a risk-stratification score is essential to help plan procedures.¹ Kancharla et al.⁸ developed a risk-stratification tool and prospectively applied this to patients undergoing TLE in a high or intermediate-risk environment. Overall 187 patients were enrolled (349 leads removed over 27 months) with similar clinical success between groups (P = 0.16) but a higher incidence of major complications in the high-risk group (6.9% vs. 0.0%; P = 0.007) and similar rates of in-hospital mortality (8.3 vs. 2.6%; P = 0.09). This study was limited by its small size and single-centre experience which can affect procedural outcomes and limit generalizability. In this article, a risk score was applied to the ELECTRa registry (3510 patients with varied operator/centre experience) and showed patients with EROS 3 were more likely to suffer significant procedure-related major complications including deaths and all-cause in-hospital major complications. Furthermore, patients in the highest risk category of EROS 3 were more than four times likely to suffer intra-procedural-related major complications including deaths, cardiac avulsion or tear, and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair. This finding is crucial as major vascular injuries require prompt surgical intervention to prevent procedural mortality and identifying these high-risk patients who are most likely to need urgent surgical intervention is critical. Notably, patients at intermediate risk (EROS 2) compared with low risk (EROS 1) were more likely to suffer all-cause in-hospital major complications including deaths, driven by heart failure, sepsis, and deaths. These excess non-procedural-related complications and deaths may also be explained by the significantly higher incidence of systemic infections in EROS 2 patients which is known to result in prolonged hospital admissions and worse long-term outcomes.¹² This suggests these patients should be closely monitored post-procedure to ensure any complications are managed effectively.

Procedural outcomes are related to patient and lead characteristics, operator experience and access to urgent surgical intervention. An analysis of the ELECTRa registry showed cases that had a higher rate of complications had both a cardiologist and surgeon present, used general anaesthesia and arterial lines.⁶ This suggests there is some appropriate risk stratification, although the similar rates of procedure-related major complications including deaths in high- and low-risk settings implies a more robust assessment is required. In the current manuscript, we have shown EROS 3 were more likely to have procedures in a high-risk environment. They were also more likely to suffer procedure-related major complications including deaths in an HiV centre and high-risk environment but not in an LoV centre and high-risk environment. Overall, despite EROS 3 having a significantly higher proportion of cardiac avulsion or tear and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair, there were similar rates of procedural-related deaths, especially intra-procedural deaths compared to lower EROS categories. This suggests when these high-risk procedures are undertaken in the appropriate environment that urgent/timely surgical intervention enables these life-threatening complications to be managed efficiently.

Application of EROS

The current risk scoring system may help risk-stratify patients undergoing TLE to ensure complications are managed as efficiently as possible. We would propose a ‘traffic light system’ to risk stratify patients into low (EROS 1: green), intermediate (EROS 2: amber), and high risk (EROS 3: red). We propose the highest risk patients (EROS 3) should be considered to have their TLE performed in a high-risk environment (operating room/hybrid theatre) environment with immediate surgical assistance available, with a cardiac surgeon present in procedures. Patients with EROS 2 are at intermediate risk of complications could have TLE performed in an environment with formalized surgical back-up (nominated cardiac surgeon available to perform thoracotomy/sternotomy but not present during the procedure). EROS 1 patients may be suitable for a low-risk environment however surgical back-up is still required. The decision of where to perform EROS 1 and 2 should also take into account frailty, additional important co-morbidities and adverse lead characteristics such as dual-coil implantable cardioverter-defibrillators that are associated with an increased risk of procedural-related complications.^4,10 This risk-stratification tool is especially important during the current pandemic when availability of general anaesthesia and surgical theatres is tightly controlled. Although this risk score identified high-risk (EROS 3) patients needing urgent surgical intervention, the finding of relatively high non-procedural-related mortality in the intermediate-risk group underlines the need to be vigilant for non-procedural-related complication related to underlying sepsis, heart failure and other co-morbidities. Additionally, EROS 3 were non-significantly more likely to suffer vascular avulsion or tears and temporary stabilization tools such as vascular occlusion balloons¹³ should be considered in these patients.

Limitations

This risk score was retrospectively applied to the ELECTRa registry and may introduce bias. However, this score was developed by analysing a previous risk stratification score and not by analysing the ELECTRa registry to identify associations between procedural outcomes and baseline patient demographics. Additionally, given the generalizability of this score depends on its applicability to a large patient cohort with a combination of HiV and LoV centres, it would be impractical to perform a prospective or randomized study of this size. There may be information bias due to the design of the study; however, dedicated data monitors oversaw data entry to ensure the accuracy of data. Participation in the registry was voluntary and not all centres in every European country were recruited which could have led to considerable inclusion bias. We considered the catheterization laboratory a low-risk environment and operating room/hybrid theatre a high-risk environment but we were unable to report on the availability of cardiothoracic assistance or tools required to manage major complications in different environments, which may have affected procedural outcomes. It should be noted that some centres performed all TLE’s in the operating room and under general anaesthesia, which would have affected patient outcomes. Overall, 122/372 (32.8%) EROS 3 patients did not have their procedures performed in a high-risk environment and this have affected how complications were dealt with potentially influencing results. EROS was developed from a study with a smaller group of patients and this is a relative weakness of the current study. Although EROS helps with risk stratification, the tools and environment required for the intervention will largely be affected by the centre availability and may be affected by additional factors such as familiarity with different tools. Major complications including pericardiocentesis, chest tube, and surgical repair were reported in conjunction and analysis of individual outcomes was outside the scope of this paper. The Kancharla et al.⁸ risk score was used to develop EROS but its appreciated there are other scores available to risk-stratify procedures but have not currently been validated in the ELECTRa registry.

Conclusions

Identifying patients at an increased risk of major complications following TLE is essential to ensure procedures are carried out with appropriate personnel and in an environment that will facilitate urgent surgical intervention. EROS was able to successfully predict patients at an increased risk of procedure-related major complications including deaths, cardiac avulsion or tear, cardiovascular lesions requiring pericardiocentesis, chest tube or surgical drain, and all-cause in-hospital major complications including deaths. EROS may provide a clinically important tool to stratify TLE patients at an increased risk of significant procedural and in-hospital complications.

Supplementary material

Supplementary material is available at Europace online.

Acknowledgements

EORP Oversight Committee, Registry Executive Committee of the EURObservational Research Programme (EORP). Data collection was conducted by the EORP department from the ESC by Myriam Glemot as Project Officer, Maryna Andarala as Data Manager. Overall activities were coordinated and supervised by Doctor Aldo P. Maggioni (EORP Scientific Coordinator).

Funding

The following companies have supported the study: Boston Scientific, Cook Medical, Medtronic, Spectranetics and Zoll. The study was supported by the Wellcome/EPSRC Centre for Medical Engineering [WT203148/Z/16/Z]. The research was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

Conflict of interest: B.S.S. is funded by NIHR and received speaker fees from EBR systems, outside of submitted work. S.A. was funded/supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. J.G. has received project funding from Rosetrees Trust, outside of submitted work. J.G., M.K.E., and V.M. have received fellowship funding from Abbott, outside of the submitted work. C.K. has presented on behalf of, advised and performed studies with Spectranetics/Philips, outside of the submitted work. J.-C.D. has received minor honoraria from Philips for lectures and consulting, outside of submitted work. A.P.M. reports personal fees from Bayer, Fresenius and Novartis, outside the submitted work. A.A. is a consultant to Boston Scientific, Backbeat, Biosense Webster, Cardiac, Corvia, Daiichi-Sankyo, EBR Systems, Medtronic, Merit, Microport CRM, Philips, and V-Wave; he received speakers’ fee from Daiichi-Sankyo, Boston Scientific, Biosense Webster, Medtronic, Microport CRM, and Philips; he participated in clinical trials sponsored by Boston Scientific, EBR Systems, Philips; he reports intellectual properties with Boston Scientific, Biosense Webster, and Microport CRM. Outside of the submitted work. K.-H.K. reports research grants (moderate) from Biosense Webster, Impulse Dynamics, and Biotronik, outside of the submitted work. C.A.R. receives research funding and/or consultation fees from Abbott, Medtronic, Boston Scientific and MicroPort, outside of the submitted work. All remaining authors have declared no conflicts of interest.

Data availability

confidentialityThe data underlying this article cannot be shared publicly due to confidentiality.

Author notes

†

Listed in Supplementary material online, Appendix S1.