Pediatric Cardiology

, Volume 34, Issue 7, pp 1620–1627

Implanted Defibrillators in Young Hypertrophic Cardiomyopathy Patients: A Multicenter Study

Authors

    • MN150 Chandler Medical CenterUniversity of Kentucky
  • Nicholas H. Von Bergen
    • UI Children’s Hospitals Pediatric Specialty ClinicsUniversity of Iowa
  • Charles A. Henrikson
    • Johns Hopkins
    • Division of Cardiovascular Medicine, MC UHN-62Oregon Health & Science University
  • Majd Makhoul
    • Sibley Heart Center CardiologyEmory University
  • Elizabeth V. Saarel
    • University of Utah
  • Martin J. LaPage
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
  • Mark W. Russell
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
  • Margaret Strieper
    • Sibley Heart Center CardiologyEmory University
  • Sunkyung Yu
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
  • Macdonald Dick
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
  • Sharlene M. Day
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
    • Cardiovascular Center 1500 East Medical Center Drive
  • David J. Bradley
    • CS Mott Children’s Hospital Floor 11 Room 661University of Michigan
Original Article

DOI: 10.1007/s00246-013-0676-6

Cite this article as:
Kamp, A.N., Von Bergen, N.H., Henrikson, C.A. et al. Pediatr Cardiol (2013) 34: 1620. doi:10.1007/s00246-013-0676-6
  • 430 Views

Abstract

Hypertrophic cardiomyopathy (HCM) is the most common genetic cardiovascular disease, with an annual risk of sudden cardiac death (SCD) estimated at 1 %. Limited data are available regarding both the risk of SCD in the young HCM population and the use of implantable cardioverter-defibrillators (ICDs). This retrospective study included all patients with HCM who underwent ICD implantation for primary or secondary prevention of SCD before the age of 30 years at five institutions between 1995 and 2009. There were 99 devices implanted in 73 patients. Appropriate shocks occurred for 11 % of all the patients. None of the previously identified conventional risk factors for SCD in HCM patients were associated with increased risk of appropriate shocks in the young study cohort. During a median follow-up period of 2.4 years, inappropriate shocks occurred for 22 % of the patients. Older age at implant was associated with a decreased risk of inappropriate shock. Those who underwent implantation in the earlier decade had a higher incidence of inappropriate shocks. Late complications including lead fracture or dislodgement, generator malfunction, and infection occurred for 32 % of the patients. Three patients died (4 %), one of whom had an arrhythmic sudden death. A greater proportion of primary prevention implantations was performed for patients from the latter decade. Over time, ICD use in young HCM patients has become increasingly primary prevention oriented. Shock rates mirror those reported in adult series, and there is a substantial incidence of device complications.

Keywords

Hypertrophic cardiomyopathyImplantable cardioverter-defibrillatorPediatricYoung adult

Hypertrophic cardiomyopathy (HCM), the most common genetic cardiovascular disease [6, 13] with a prevalence of 1:500, is the most common cause of sudden cardiac death (SCD) in young people [14]. The annual risk of SCD is estimated to be approximately 1 % in the general HCM population, although higher rates are reported from tertiary centers [14].

Clinical risk factors associated with SCD in adult HCM patients have emerged including maximal left ventricular (LV) wall thickness greater than 30 mm [22], nonsustained ventricular tachycardia [16], abnormal blood pressure response to exercise [20], unexplained syncope [23], and family history of SCD [14]. In fact, the presence of only one of these risk factors may justify an implantable cardioverter-defibrillator (ICD) for primary prevention of SCD [6, 15, 25].

More recently, delayed enhancement on cardiac magnetic resonance imaging (CMRI), a marker of myocardial fibrosis, has emerged as another potential risk factor for SCD [1, 9, 24]. Additionally, the increasing availability of genetic testing has allowed its use as a diagnostic tool [12, 17, 28].

Although ICDs in HCM patients result in appropriate shock rates of 2.3–4 % per year [15, 18, 25], in 30–35 % of patients they also are associated with implant complications, inappropriate shocks, or other device complications [11, 18]. Much of the current data regarding ICD use in HCM patients is based on adult patients, and the risk factors associated with SCD in adults with HCM have not been validated in younger patients. Furthermore, limited data regarding indications for ICD placement and associated appropriate and inappropriate shocks have been reported for young patients with HCM [19]. Consequently, the role of ICD placement for primary prevention of SCD in young HCM patients is unclear. We report a group of all the HCM patients from five centers who underwent ICD implantation at an age younger than 30 years.

Methods

Patient Inclusion

All patients with HCM who underwent ICD implantation before the age of 30 years were identified at five academic institutions. The patient diagnosis was made at the participating centers based on clinical and imaging data. Patients were excluded if they had other structural heart disease, genetic syndromes, or metabolic syndromes. Institutional Review Board approval was obtained from all the participating centers. Patient consent was waived due to the retrospective study design.

Patient data were obtained from a retrospective medical review at each participating center. All medical records were de-identified at submission to the coordinating center.

Risk Factors

Data collection was performed using a uniform data collection tool designed on the basis of previously identified risk factors in primarily adult HCM cohorts including LV wall thickness greater than 30 mm, unexplained syncope, abnormal blood pressure response to exercise, nonsustained ventricular tachycardia observed on Holter, and family history of SCD. Because these risk factors are used in current risk assessment of SCD in HCM patients, this report refers to them collectively as conventional risk factors.

Data collection also included patient history, family history, device implantation and interrogation data, clinical findings, and genetic data. Implant complications were defined as early (occurring within 30 days after implantation) or late (occurring more than 30 days from implantation) device complications. Complications included lead fracture, device recall, infection requiring antibiotics or explantation, generator malfunction, or lead dislodgement. Follow-up data were obtained through January 2011.

Device electrogram evaluation and classification of arrhythmias were performed at each contributing center by an electrophysiologist. Appropriate shocks were defined as any shock triggered by ventricular fibrillation or ventricular tachycardia. Inappropriate shocks were defined as any shock triggered by other rhythms or device/lead malfunction.

Statistical Analyses

Statistical analyses were performed at the coordinating center. Demographic, clinical, and echocardiographic characteristics of primary and secondary prevention groups and those between the first decade (1990–2000) and second (2001–2010) decade were compared using Fisher’s exact test for categorical variables and t tests or the Wilcoxon rank-sum test for continuous variables.

The generalized estimating equations method [10] was used to determine whether a difference in inappropriate shocks existed between the device manufacturers. The cumulative incidence rate of appropriate or inappropriate shocks since ICD implantation was computed by the Kaplan–Meier method, and patients with primary and secondary prevention implants were compared using the log-rank test.

The uni- and multivariate Cox regression model was used to identify possible risk factors associated with appropriate or inappropriate shocks. The results of the Cox regression model were presented as hazard ratios. All analyses were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA). Statistical significance was set at p values lower than 0.05 using two-sided tests.

Results

Patient Characteristics

Between 1995 and 2009, 99 devices were implanted in 73 patients, with a median follow-up period of 2.4 years (range, 0.1–12.9 years). The patient characteristics are summarized in Table 1. The absolute septal thickness was significantly greater in the primary prevention group than in the secondary prevention group. No difference in prevalence of the other conventional risk factors was observed between the primary and secondary prevention groups. Seven patients (11 %) in the primary prevention group met none of the conventional risk factors as indications for ICD implantation.
Table 1

Demographic variables and risk factors

 

All

Primary prevention

Secondary prevention

P value

(n = 73)

(n = 61)

(n = 12)

n (%)

n (%)

N (%)

Demographics

 Mean age at implant (years)

14.8 ± 4.9

15.1 ± 5.0

13.1 ± 4.7

0.19

 Median follow-up: years (range)

2.4 (0.1–12.9)

2.6 (0.2–11.4)

1.4 (0.1–12.9)

0.13

 Males

43 (59)

34 (56)

9 (75)

0.34

 History of myectomy

14 (19)

12 (20)

2 (17)

1.0

Risk factors

 Family history of SCD

24 (33)

20 (33)

4 (33)

1.0

 Family history of HCM

42 (58)

35 (57)

7 (58)

 

 Unexplained syncope

12 (16)

10 (16)

2 (17)

1.0

 Exercise test performed

26 (36)

26 (43)

0 (0)

 

  Abnormal BP response

11 (42)

11 (42)

 

NA

 Spontaneous VT

10 (14)

7 (11)

3 (25)

0.36

Pre-ICD Holter

 

1.0

 Normal, PVCs (single/couplets)

37 (51)

32 (52)

5 (42)

 

 Triplets, NSVT, VT

9 (12)

8 (13)

1 (8)

 

 Not performed/unknown

27 (37)

21(34)

6 (50)

 

Type of hypertrophy, echo

 Asymmetric septal

57 (78)

50 (82)

7 (58)

0.20

 Concentric

11 (15)

8 (13)

3 (25)

 

 Apical

2 (3)

1 (2)

1 (8)

 

 Unknown

3 (4)

2 (3)

1 (8)

 

Asymmetric septal hypertrophy

 Mean septal thickness ED (mm)

25.5 ± 8.6

26.4 ± 8.2

17.8 ± 7.8

0.02

 Mean septal thickness ED (z-score)

11.3 ± 7.1

12.1 ± 6.9

7.4 ± 7.3

0.06

LV wall thickness (mm)

 

0.27

 <30

49 (67)

40 (66)

9 (75)

 

 ≥30

18 (25)

17 (28)

1 (8)

 

 Unknown

6 (8)

4 (6)

2 (17)

 

Genetic testing performed

25 (34)

20 (33)

5 (42)

0.50

 Positive

19 (76)

16 (80)

3 (60)

0.56

MRI performed

24 (33)

21 (33)

3 (25)

 

 Mean LV mass (g/m2)

130.6 ± 40.1

137.6 ± 39

88.5 ± 4.9

0.11

 Delayed enhancement

18 (75)

17 (81)

1 (33)

0.14

Device data

 Dual-chamber system

44 (60)

37 (61)

7 (58)

1.0

Lead placement

 Epicardial

3 (4)

1 (2)

2 (17)

0.07

 Transvenous

70 (96)

60 (98)

10 (83)

 

No. of devices

 1

53 (73)

43 (70)

10 (83)

0.49

 2

14 (19)

13 (21)

1 (8)

 

 3

6 (8)

5 (8)

1 (8)

 

SCD sudden cardiac death, HCM hypertrophic cardiomyopathy, BP blood pressure, NA not applicable, VT ventricular tachycardia, ICD implantable cardioverter-defibrillator; PVC premature ventricular contractions, NSVT nonsustained ventricular tachycardia, ED end-diastolic, LV left ventricle, MRI magnetic resonance imaging

The number of patients undergoing CMRI did not differ between the primary and secondary prevention groups. In 25 patients (34 %) who had genetic testing performed, known mutations were identified in 19 patients who had genetic testing performed. Five of these patients (25 %) had an MYH7 gene mutation, and six patients (32 %) had a mutation in the MYBPC3 gene. Subgroup analysis of the patients younger than 18 years was not performed.

Outcomes

Most of the devices (64 %) were programmed with one therapy zone, and 38 % of the implanted devices had a monitor zone. Inappropriate shocks did not differ between the device manufacturers. Appropriate shocks occurred for eight patients (11 %) (Table 2), and median time to the first appropriate shock was 1.7 years (range, 0.2–7.0 years). The rate of freedom from appropriate shocks in the primary prevention group was significantly higher than in the secondary prevention group (p = 0.03; Fig. 1). The incidence of appropriate shocks was 2.8 % per 100 patient-years.
Table 2

Outcomes

 

All

Primary prevention

Secondary prevention

P value

(n = 73)

(n = 61)

(n = 12)

n (%)

n (%)

n (%)

Appropriate shocks

8 (11)

6 (10)

2 (17)

0.61

Appropriate ATP

3 (4)

2 (3)

1 (8)

0.42

Inappropriate shocks

16 (22)

15 (25)

1 (8)

0.28

Early implant complications

 (≤30 days of implant)

9 (12)

8 (13)

1 (8)

1.0

Late device complications

23 (32)

20 (33)

3 (25)

0.74

 Fidelis lead recall

11

   

 Device recall

6

   

 Lead fracture

4

   

Required generator replacement

20 (27)

18 (30)

2 (17)

0.49

Required lead replacement

14 (19)

13 (21)

1 (8)

0.44

Death

3 (4)

2 (3)

1 (8)

0.42

Transplant

2 (3)

1 (2)

1 (8)

0.30

ATP anti-tachycardia pacing

https://static-content.springer.com/image/art%3A10.1007%2Fs00246-013-0676-6/MediaObjects/246_2013_676_Fig1_HTML.gif
Fig. 1

Appropriate shocks. Freedom from appropriate shocks is plotted over years of follow-up evaluation. Primary prevention implants are plotted in blue, and secondary prevention implants are plotted in red. The median time to the first appropriate shock was 1.7 years. The incidence of appropriate shocks was 2.8 % per 100 patient-years

Appropriate anti-tachycardia pacing therapy was delivered successfully for three patients (4 %). The rate of appropriate shocks did not decrease in those patients taking class I, II, or III anti-arrhythmic medications.

Inappropriate shocks occurred for 16 patients (22 %). The time to the first inappropriate shock was 1.8 years (range, 0.02–8.8 years). The rate of freedom from inappropriate shocks did not differ between the primary and secondary prevention groups (Fig. 2). The incidence of inappropriate shocks was 5.7 % per 100 patient-years.
https://static-content.springer.com/image/art%3A10.1007%2Fs00246-013-0676-6/MediaObjects/246_2013_676_Fig2_HTML.gif
Fig. 2

Inappropriate shocks. Freedom from inappropriate shocks is plotted over years of follow-up evaluation. Primary prevention implants are plotted in blue, and secondary prevention implants are plotted in red. The median time to the first inappropriate shock was 1.8 years. The incidence of inappropriate shocks was 5.7 % per 100 patient-years

The primary and secondary prevention groups did not differ with regard to the incidence of early implant complications (9 patients, 12 %) or late device complications (23 patients, 32 %). Late complications due to the Fidelis lead occurred in 11 patients (48 %) and included lead replaced electively or lead malfunction.

The cohort had three deaths (4 %). One boy with a history of multiple inappropriate shocks died after ventricular fibrillation occurred below his detection zone. One subject died of bacteremia and sepsis at an outside hospital less than 60 days after implantation. The cause of death for one subject lost to follow-up evaluation is unknown.

Appropriate Shocks

None of the previously identified conventional risk factors for SCD in HCM patients were significantly associated with increased risk of appropriate shocks in our young cohort (Table 3). No patients who received appropriate shocks had ventricular triplets or nonsustained ventricular tachycardia observed on Holter monitoring. Univariate analyses showed that patients who were older at the time of implantation had a decreased risk of appropriate shock. Family history of SCD was associated with an increased risk of appropriate shock, with a trend toward significance.
Table 3

Risk factors for appropriate shocks

Variable

HR

95 % CI of HR (lower–upper)

P value

Univariate analyses

 Male

2.6

(0.5–13.1)

0.22

 Older age at implant

0.8

(0.7–0.98)

0.03

 Syncope, unexplained

1.2

(0.3–6.3)

0.79

 Abnormal BP with exercise test

2.3

(0.1–40.4)

0.57

 Family history of SCD

4.4

(0.8–22.5)

0.06

 Echo data

  Asymmetric hypertrophy

1.1

(0.2–5.7)

0.91

  LV wall thickness >30 mm

1.3

(0.2–6.5)

0.78

 ICD system

  Single

3.6

(0.8–15.6)

0.07

Multivariate analyses

 Family history of SCD

10.6

(1.2–90.2)

0.03

 Abnormal BP response to exercise or abnormal Holter or positive genetic testing

6.2

(0.9–40.5)

0.06

 History of resuscitated arrest

4.5

(0.8–26.1)

0.09

 Syncope, unexplained

7.7

(0.7–80.7)

0.09

HR hazard ratio, CI confidence interval, BP blood pressure, SCD sudden cardiac death, LV left ventricle, ICD implantable cardioverter-defibrillator

Because the sample size was limited, only select variables were considered in the multivariate analysis. Additionally, because not every patient’s pre-implantation evaluation had all the variables of interest, the three following clinical variables were combined as a single variable in the multivariate analysis: abnormal blood pressure response to exercise, abnormal Holter, and positive genetic testing. A trend toward significant association was found between appropriate shock and this composite risk factor, history of resuscitated SCD, and unexplained syncope. According to the multivariate analysis, family history of SCD was significantly associated with increased risk of appropriate shock.

The proportion of patients with a history of appropriate shocks during the study period was 11 % (8/73). With a two-sided 0.05 significance level and a power of 80 %, a sample size of at least 263 patients would be required to detect a hazard ratio of 3.0 with this incidence of appropriate shocks.

Inappropriate Shocks

The rate of inappropriate shocks did not differ between implanted single- and dual-chamber systems (Table 4). Atrial fibrillation (AF) was associated with an increased risk of inappropriate shocks. Our data collection did not identify the time of AF diagnosis. Therefore, the diagnosis of AF may have been made at the time of inappropriate shock. Multivariate analysis showed that even in this young cohort, older age at implantation was associated with decreased risk of inappropriate shock.
Table 4

Risk factors for inappropriate shocks

Variable

HR

95 % CI of HR (lower–upper)

P value

Univariate analyses

 Male

2.2

(0.7–6.5)

0.15

 Older age at implant

 ≥13

0.5

(0.2–1.4)

0.21

 ICD system

 Single

0.6

(0.2–1.9)

0.39

 Atrial fibrillation

4.8

(1.3–17.4)

0.01

Multivariate analyses

 Older age at implant

0.8

(0.7–0.9)

0.003

 ICD system

 Single

0.5

(0.2–1.8)

0.33

 Atrial fibrillation

39.8

(5.3–300.1)

<0.01

HR hazard ratio, CI confidence interval, ICD implantable cardioverter-defibrillator

Decade of Implant

The cohort was divided based on the date of initial implantation falling into the first decade (1990–2000) or second (2001–2010) decade represented in the series (Table 5). No difference in age at implantation, gender, or any of the conventional risk factors was observed between those implanted in the earlier group and those implanted in the more recent decade. A greater proportion of primary prevention implantation was performed in the latter decade. The two groups did not differ in the incidence of appropriate shocks, but those implanted in the earlier decade had a higher incidence of inappropriate shocks, with no difference in the time to the first inappropriate shock. Additionally, those implanted earlier had a higher incidence of reoperation than those implanted later, with no difference in the time to reoperation.
Table 5

Period of implant

 

1990–2000

2001–2010

P value

(n = 10)

(n = 63)

n (%)

n (%)

Mean age at implant (years)

12.4 ± 5.4

15.2 ± 4.8

0.10

Male

5 (50)

38 (60)

0.73

Implant indication

 Primary prevention

6 (60)

55 (87)

0.05

 Secondary prevention

4 (40)

8 (13)

 

Median follow-up: years (range)

10.2 (0.1–12.9)

2.3 (0.2–9.4)

0.001

Outcomes

 Appropriate shocks

2 (20)

6 (10)

0.30

 Inappropriate shocks

5 (50)

11 (17)

0.04

 Median time to IS: years (range)

1.7 (0.1–8.8)

1.9 (0.02–8.1)

0.62

 Early implant complications (≤30 days)

3 (30)

6 (10)

0.08

 Late device complications

4 (40)

19 (30)

0.72

 Required lead replacement

3 (30)

11 (17)

0.39

 Reoperation

8 (80)

12 (19)

0.0003

 Median time to reoperation: years (range)

4.4 (0.9–8.9)

5.0 (2.0–7.6)

0.50

 Death

1 (10)

2 (3.2)

0.36

IS inappropriate shock

Discussion

Previous reports of risk factors for SCD in HCM adults have included a broad age range of patients, with a mean age in the fifth decade of life. The population in this study included only children and young adults with a mean age of 14.8 years. In these patients, implantation indications mirrored those reported in adult registry data, suggesting that pediatric physicians are adopting recommendations from adult data regarding primary prevention implantation. However, seven patients implanted for primary prevention (11 %) met none of the conventional risk factors for primary prevention implantation. Some of these patients were implanted in an earlier period before identification of current risk factors, and the potential presence of current risk factors was not reported. Additionally, the majority of these patients were implanted under age 15 years, and we hypothesize that providers used other data such as delayed enhancement on MRI and genetic results in their risk assessment.

The rate of appropriate shocks in our study was similar to that reported from adult registry data, emphasizing the life-saving potential of these devices in HCM. However, the predictive value of individual conventional risk factors was likely influenced by the sample size of the cohort. Additionally, because 67 % of our primary prevention implants with appropriate shocks had only one conventional risk factor before implantation, the presence of one such risk factor for SCD in young HCM patients may similarly justify the implantation of an ICD for primary prevention.

The incidence of inappropriate shocks in our study is consistent with previously reported multicenter data regarding the use of ICDs in pediatric and congenital heart disease patients [4, 26] as well as adult registry data of ICD use in HCM [3, 15]. Many inappropriate shocks in our cohort were related to the failure of a single lead model (Sprint Fidelis; Medtronic Corp., Minneapolis MN, USA), and we expect this number to decrease in future years. As previously reported in the pediatric literature [8]; no reduction in inappropriate shocks was conferred by dual-chamber systems.

Data emerging in recent years support primary prevention implantation in both ischemic and nonischemic cardiomyopathy [2, 7], and implantation trends have shifted toward increased primary prevention implantations [4, 21]. A similar shift also has occurred for prevention of SCD in HCM. Primary prevention devices represented nearly a 50 % greater proportion of total devices in the second decade of the series than in the first decade. The rate of appropriate shocks did not differ significantly in the more recent decade, suggesting that the expanded use of ICDs for primary prevention has not led to excessive or inappropriate implantations. The higher incidence of inappropriate shocks in the earlier decade of implantation, with no difference in the time to the first inappropriate shock, suggests that programming and technology have improved over time.

The use of ICDs in children and young adults for any reason is known to be associated with increased implant complications compared with the adult population [5, 27]. The high complication rate of 40 % in the earlier decade did not decline for patients in the latter decade, although the reduced incidence of early complications and inappropriate shocks with more recent devices is encouraging. Finally, although the appropriate shock rate (11 %) demonstrates a reasonable number of patients benefiting from their devices, ICDs did not prevent all arrhythmia-related deaths.

The results of this study are limited by the small number of patients. Power analysis predicts that the study population would need to be more than double to detect a hazard ratio of 4.0 for appropriate shocks.

Conclusion

The use of ICDs in young HCM patients has become predominantly oriented toward primary prevention, with appropriate and inappropriate shock rates that mirror those reported in adult series. Although the conventional risk factors identified from adult data were not associated with an increased risk of appropriate shocks in this younger study population, the presence of only one risk factor may similarly justify ICD implantation for primary prevention of SCD. In this clinical scenario, ICDs remain life-saving tools, but their use places a heavy burden on patients due to their associated complications.

Copyright information

© Springer Science+Business Media New York 2013