Received: 23 November 2016 Revised: 25 July 2017 Accepted: 14 August 2017

DOI: 10.1111/pace.13197

DEVICES

How active are young cardiac device patients? Objective

assessment of activity in children with cardiac devices

Caridad M. de la Uz MD1 Ashley E. Burch PhD2 Bruce Gunderson MS3

Jodi Koehler MS3 Samuel F. Sears PhD4

1 The Lillie Frank Abercrombie Section of Pedi-

atric Cardiology Texas Children's Hospital, Abstract

Baylor College of Medicine, Houston, TX, USA
2 Department of Psychology, East Carolina
University, Greenville, NC, USA
3 Medtronic, Inc., Minneapolis, MN, USA
4 Department of Psychology and Department
of Cardiovascular Sciences, East Carolina
University, Greenville, NC, USA
Correspondence
Samuel F. Sears, PhD, Department of Psychol-
ogy, East Carolina University, 104 Rawl Bldg,
Greenville, NC 27858.
Email: searss@ecu.edu
Background: The daily activity of pediatric patients with implantable cardiac devices provides
behavioral evidence of functional outcomes. Modern devices provide continuous accelerometer
data that are sensitive to movement, but normative values have not been published for pediatric
activity rates. This study provides the first normative accelerometer data on activity rates in a
large sample of pediatric cardiac device patients.
Methods: Patients were between 3 and 18 years old (N = 1,905) and implanted with a cardiac
device from a single device company, and enrolled in remote monitoring.
Results: The median age at implant was 14 years (interquartile range = 12–16); 61.3% were male.
Data for 4 weeks were extracted from a company database at 53 weeks postimplant and an aver-
age of daily activity was calculated. Daily average activity for all patients was 5.4 hours (standard
deviation = 2.0). In a multivariate analysis, increased level of activity was associated with: being
male, having a pacemaker versus implantable cardioverter defibrillator (ICD), epicardial device
location, rate response turned off, having experienced a shock, and younger age.

Conclusions: These results provide the first baseline data of physical activity in children with
implanted cardiac devices and provide a clinical guide to physical activity assessment in this popu-
lation. Further, our data suggest physical activity in children with implantable cardiac devices may
differ based on demographic variables, device type, device location, indication for implantation,
and history of ICD shock.

KEYWORDS
ICD, implantable cardiac device, pediatric, physical activity, PM

1 INTRODUCTION
Physical activity in childhood and across the lifespan is associated
with broad health outcomes1–3 and its importance cannot be under-
stated. Studies of activity in large pediatric samples have shown phys-
ical activity decreases as children age.4,5 Children with compromised
cardiac functioning, particularly those with an implantable cardiac
device, may be at an even greater risk for reduced physical activity
due to elevated levels of anxiety and activity avoidance.6,7 Two small
qualitative studies of children with implantable cardioverter defib-
rillators (ICDs) found patients reported diminished physical activity
compared to activity level prior to implant.8,9 Most studies evaluat-
ing activity in cardiac device patients utilize self-reported measures
of physical activity. However, for patients with implantable cardiac
devices obtaining an objective measure of physical activity is possi-
ble since many modern implantable cardiac devices have an integrated
accelerometer.
Studies examining health outcomes in patients with implanted car-
diac devices have found factors such as gender, device type, and
age to be associated with variations in activity and physical quality
of life. In adult cardiac device patients, females experience greater
overall distress and lower physical functioning when compared to
males.10,11 In a pediatric sample, female ICD patients reported lower
physical quality of life scores than males.7 Device type also differen-
tially affects quality of life and anxiety. Compared to patients with a
pacemaker (PM), children with ICDs report reduced quality of life12
and increased anxiety.13 However, when Webster et al.13 accounted
for age at implant the relationship between device type and anxiety
was no longer significant; instead, patients’ age at the time of implant
was related to level of anxiety, with older patients reporting greater

Pacing Clin Electrophysiol. 2017;40:1286–1290. wileyonlinelibrary.com/journal/pace 

c 2017 Wiley Periodicals, Inc. 1286
DE LA UZ ET AL . 1287

anxiety than patients implanted at a younger age. It is plausible that is comparable to walking approximately 70 steps/minute.16 For each
anxiety and activity avoidance increase as children become aware of participant, the number of minutes active per day is stored in the
their cardiac condition, the risk of damaging the device during activ- cardiac device for 425 days. Queries and analyses were performed
ity, and the potential for shock from an ICD, and as a result, limit their using SAS (v9.4) and JMP (v11.2.0) software (SAS Institute Inc., Cary,
physical activity. How the proposed decline in activity among pediatric NC, USA); two-tailed alphas of P < 0.05 were interpreted as significant.
cardiac device patients correlates with the decrease in physical activity
with age observed in healthy pediatric populations is unknown.
Objective measurement of activity rates in pediatric patients 2.3 Statistical analysis
with implantable cardiac devices has not been studied. The primary
Analyses were performed to ascertain differences in daily physical
purpose of this study is to provide the first normative data on activity
activity level in children based on age at time of implant. A 4-week
rates in pediatric cardiac device patients. Our secondary purpose is to
(28 days) sample of activity data at the 53 weeks postimplant time
examine the effect of gender, device type and characteristics, implant
point was extracted from the Medtronic CareLink database. Daily
indication, ICD shocks, and age at implant on physical activity. It is
average activity was calculated by summing the number of active hours
hypothesized that these variables will influence activity rates as fol-
across 28 consecutive days and dividing by 28. Age at time of implant
lows. As far as demographics are concerned, it is hypothesized that
and overall activity levels are reported as means and standard devi-
children with cardiac devices would follow the same trends in activity
ations (SD). Gender, device type and characteristics, indication, and
level as their nonpaced colleagues in the general population. Extant lit-
shock among the sample are given as percentages.
erature reveals that boys are more active than girls14 and that activ-
Activity differences were investigated using a univariate approach
ity decreases with age in children. We expect that the younger group
with individual predictors. Differences in average activity based on
(3–11 years) will be more active than the older group (12–18 years).
gender, device type, rate response, device location, and shock were
Patients with ICDs will exhibit reduced physical activity compared to
assessed using t-tests. Age was dichotomized and a t-test was used to
PM patients, as has been the case in the adult literature and also
examine difference in average activity between the younger group (3–
because children with ICDs report a higher degree of anxiety and self-
11 years) and the older group (12–18 years). An analysis of variance
limit activity. We predict that those with rate-responsive pacing will be
was used to detect difference in activity based on number of chambers
more active than those without rate responsive pacing. The adult liter-
paced and indication. A multivariate regression using a stepwise entry
ature shows improved exercise capacity with rate-responsive pacing,15
method with selection based on the Schwarz Bayesian information cri-
so we predict that improved capacity to exercise may lead to increased
terion was used to identify independent predictors of activity.
activity. Patients with epicardial pacing systems will have lower activity
levels. Patients with indications for implant including sudden cardiac
death/ventricular tachycardia and cardiomyopathy will also have lower
activity levels, given their self-imposed or disease-imposed activity 3 RESULTS
restrictions. We also predict that patients receiving at least one shock
may have lower activity levels than those who do not receive shocks, A total of 1,905 patients met the criteria and were included in the
since this has been shown in the adult literature to be true. analyses. The median age at implant was 14 years (interquartile
range = 12–16); 61.3% were male. The majority of patients had an ICD
(n = 1,637, 85.9%); 268 (14.1%) had a PM.
2 METHODS Daily average activity for all patients was 5.4 hours (SD = 2.0). Uni-
variate analyses reveal that on average, younger children (3–11 years)
2.1 Participants were more active than older children (12–18 years), 6.7 hours ± 2.0
versus 4.9 hours ± 1.8, respectively (P < 0.001), and males spent nearly
All patients had a single, dual, or biventricular PM or ICD. Patients were
an hour more per day being active compared to females, 5.7 hours
included in the current study if they were between 3 and 18 years
± 2.0 versus 4.8 hours ± 1.8, respectively, P < 0.001 (see Table 1).
old and enrolled in Medtronic CareLink remote monitoring for at least
Daily activity differences were also found based on device type. Signif-
1 year from implant date. Participants were excluded if they experi-
icant results showed PM patients (6.0 hours ± 2.2) had higher activ-
enced a recent shock. Specifically, patients who experienced a shock
ity levels than ICD patients (5.3 hours ± 1.9), P < 0.001. Further,
during the 4-week sample of activity, or the 30 days preceding the
greater level of activity was associated with an epicardial device loca-
4-week sample, were excluded. The date range for the current study
tion and having experienced a shock. Activity differences were not
was March 2002–October 2014.
found between groups for rate response (on/off) or among groups
based on number of chambers paced (single chamber, dual chamber,
2.2 Physical activity
or biventricular pacing) (P's > 0.2). Device indication was also signif-
Physical activity was continuously measured with single-axis icantly related to activity; patients with bradycardia/sinus node dys-
accelerometer sensor inside the cardiac device. A minute is counted function/atrial tachyarrhythmias demonstrated the highest level of
as active if a threshold, which includes the number and magnitude activity while patients with cardiomyopathy/heart failure had the low-
of deflections in the accelerometer signal, is reached. The threshold est activity rates on average.
1288 DE LA UZ ET AL .

TA B L E 1 Daily activity in hours TA B L E 2 Multivariate regression results

Mean Daily Activity

Variable (SD) (hours.) P Regression Average Change in
Predictor Estimate Minutes per Day P-value
Gender <0.001
Male 0.84 50.2 <0.001
Male 5.7 (2.0)
Pacemaker 0.64 38.5 <0.001
Female 4.8 (1.8)
Epicardial lead 0.55 33.0 <0.001
Device type <0.001
Rate response off 0.43 25.6 <0.001
ICD 5.3 (1.9)
ICD shock 0.37 22.0 0.001
PM 6.0 (2.2)
Age (continuous) −0.20 −11.9 (for each year) <0.001
Rate-response 0.219
On (n = 277) 5.2 (2.3) ICD = implantable cardioverter defibrillator

Off (n = 1,628) 5.4 (1.9)

Chamber 0.378
the daily average activity for all patients was approximately 5 hours.
Single 5.4 (1.9) Males were more active compared to females by nearly an hour per
Dual 5.3 (2.0) day. This anticipated finding of increased activity in males compared
CRT 5.5 (2.7) to females is reported widely among healthy pediatric samples.17–19
Device location <0.001 Significant differences were also found with regard to device type.
Epicardial 6.8 (2.0) Patients with a PM were more active than those with an ICD. Despite
Transvenous 5.2 (1.9) the current trend toward liberalizing activity restrictions in some
Indication <0.001 patients with ICDs, physician-imposed activity limitations remain com-

AV block 5.5 (2.2) mon practice for patients with ICDs. The leading indications for ICD
implantation in pediatric patients include cardiac dysrhythmias, car-
Bradycardia/SND/atrial 6.2 (2.1)
tachyarrhythmias diomyopathy, and channelopathy, while the leading indications for
Syncope and SCD/VT 5.3 (1.8) PM placement are heart block, congenital heart disease, and cardiac
Cardiomyopathy/heart 5.0 (2.0) dysrhythmia.12 By definition, those who undergo ICD implantation
failure have either had or are at risk of having a potentially lethal ventricular
Other 5.4 (2.0) arrhythmia, which may argue that disease severity may be greater in
Shock 0.006 the ICD group. Finally, self-imposed restrictions due to fear of appro-
Yes 5.7 (2.0) priate or inappropriate shock in ICD patients may also affect activity
No 5.3 (2.0) level as patients become more aware of their disease.7

Age <0.001
Younger 6.7 (2.0)
Older 4.9 (1.8)
AV = atrioventricular; CRT = cardiac resynchronization therapy; ICD =
implantable cardioverter defibrillator; PM = pacemaker; SCD = sudden car-
diac death; SD = standard deviation; SND = sinus node dysfunction; VT =
ventricular tachycardia
All variables were entered into the multivariate analysis. Two of the
variables, indication and number of chambers paced, did not meet the
criteria for inclusion in the final model. All other variables achieved
significance while holding other variables in the model constant. Gen-
der, device type, and device location had the largest effect on activity
(see Table 2). Being male was associated with an activity increase of
50.4 minutes per day, having a PM was associated with an activity
increase of 38.4 minutes per day, and an epicardial device was associ-
ated with an activity increase of 33 minutes per day on average.
4 DISCUSSION
The current study provided the first objective data of physical activity
in children with PMs and ICDs. Results from this study indicated that
Patients with epicardial PMs showed greater activity levels than
their counterparts with transvenous pacing systems. One possible
explanation for this is that indication for implantation was not able to
be accounted for in the multivariable analysis; therefore, patients with
epicardial pacing systems may represent a greater proportion of the
patients with less severe diseases such as congenital atrioventricular
block and sinus node dysfunction, which were shown to be associated
with higher activity levels. There is also a possibility that implant loca-
tion may have an effect on the ability of the accelerometer to detect
activity. This is an interesting question worthy of further exploration.
The presence of an ICD shock was surprisingly associated with
increased activity levels, which is the opposite of what the adult litera-
ture has shown. The nature of this study precludes us from distinguish-
ing appropriate from inappropriate shocks in this patient population.
Therefore, it is unclear whether the shocks documented were received
as appropriate therapy for a ventricular tachyarrhythmia or as an inap-
propriate shock due to oversensing or misinterpretation of a supraven-
tricular arrhythmia, including sinus tachycardia. If the latter is the case,
then it makes sense that patients who are more active are going to
be at greater risk of supraventricular tachyarrhythmias including sinus
tachycardia or oversensing. It is widely recognized that the incidence of
inappropriate shocks is greater in children than in adults. It is also plau-
sible that patients with diseases where catecholamine surges lead to
DE LA UZ ET AL .
ventricular tachyarrhythmias may be receiving appropriate therapies
when they participate in activities which can trigger their events.
The presence of rate-responsive pacing did not significantly affect
activity level. While rate-responsive pacing does improve exercise tol-
erance in those with chronotropic incompetence, this does not neces-
sarily translate to increased activity. It is also possible that a difference
was not detected because we compared activity in patients with rate-
responsive pacing to those without rate-response. This latter group
may consist of patients with chronotropic incompetence whose device
settings are not optimized for their disease, in addition to patients with
normal chronotropic responses who did not require rate-response.
Declining physical activity as a function of age has been well
described in children in the general population in the United States and
other countries and across different activity measurement modalities
(e.g., wearable accelerometers, pedometers, self-reporting).4,5,17,20
The current study supported the finding of reduced physical activity as
a function of age; younger patients (ages 3–11) demonstrated signifi-
cantly greater daily physical activity, nearly 2 hours more per day, than
their older counterparts (ages 12–18). Among the general population,
one study reported a decrease in activity of 4.2% per year after age
5.4 Beyond the potential reasons for the age-associated trend toward
inactivity in the general population, patients with cardiac implantable
devices may also be more likely to self-limit activities, as described by
Sears et al.7
The current study provides novel information but has some lim-
itations to consider. Limitations of this study include its retrospec-
tive nature and the limited access to relevant clinical data such as
indications for device implantation, severity of disease, detailed shock
history for patients with ICDs, comorbidities, anthropometric data,
and socioeconomic/demographic data. For example, some ICD patients
may carry diagnoses, such as catecholamine-sensitive polymorphic
ventricular tachycardia and other channelopathies with associated
arrhythmias, which frequently mandate activity limitations. Our study
did not attempt to decouple prescribed activity limitations from the
data because this level of information is not available in a device
company database. In addition, pediatric device patients have more
idiosyncratic implantation considerations related to anatomical or
pathophysiological differences that may affect our data versus adult
data sets. The limitations in the data follow from the trade-off for one
of the largest cardiac device, pediatric databases ever published.
The results of this study provide the first ever description of
accelerometer-determined activity in pediatric PM and ICD patients.
Further examination of existing device-based accelerometer data in
children may allow for better understanding about how to address
device- or disease-specific concerns, and incorporate activity promo-
tion into the clinical care of pediatric device patients. These data pro-
vide reference points for the clinic-based education of patients and
families related to normative activity levels.
ORCID
Ashley E. Burch PhD http://orcid.org/0000-0003-3871-6349
1289
REFERENCES
1. Janssen I, LeBlanc AG. Systematic review of the health benefits of
physical activity and fitness in school-aged children and youth. Intl J
Behav Nutr Physical Activity. 2010;7:40.
2. Jiménez-Pavón D, Konstabel K, Bergman P. Physical activity and clus-
tered cardiovascular disease risk factors in young children: a cross-
sectional study (the IDEFICS study). BMC Med. 2013;11:1.
3. Timmons BW, LeBlanc AG, Carson V. Systematic review of physical
activity and health in the early years (aged 0–4 years). Appl Physiol Nutr
Metabol. 2012;37:773–792.
4. Cooper AR, Goodman A, Page AS. Objectively measured physical
activity and sedentary time in youth: the International Children's
Accelerometry Database (ICAD). Intl J Behav Nutr Physical Activity.
2015;12:113.
5. Dumith SC, Gigante DP, Domingues MR, Kohl HW. Physical activity
change during adolescence: a systematic review and a pooled analysis.
Intl J Epidemiol. 2011;40:685–698.
6. Koopman HM, Vrijmoet-Wiersma CMJ, Langius JND. Psycholog-
ical functioning and disease-related quality of life in pediatric
patients with an implantable cardioverter defibrillator. Pediatric Car-
diol. 2012;33:569–575.
7. Sears SF, Hazelton AG, Amant JS. Quality of life in pediatric
patients with implantable cardioverter defibrillators. Am J Cardiol.
2011;107:1023–1027.
8. Rahman B, Macciocca I, Sahhar M, Kamberi S, Connell V, Dun-
can RE. Adolescents with implantable cardioverter defibrillators: a
patient and parent perspective. Pacing Clin Electrophysiol. 2012;35:62–
72.
9. Eicken A, Kolb C, Lange S. Implantable cardioverter defibrillator (ICD)
in children. Intl J Cardiol. 2006;107:30–35.
10. Walker RL, Campbell KA, Sears SF. Women and the implantable car-
dioverter defibrillator: a lifespan perspective on key psychosocial
issues. Clin Cardiol;27:543–546.
11. Starrenburg A, Pedersen S, den Broek KR, Kraaier K, Scholten M,
Palen J. Gender differences in psychological distress and quality of life
in patients with an ICD 1-year postimplant. Pacing Clin Electrophysiol.
2014;37:843–852.
12. Czosek RJ, Bonney WJ, Cassedy A. Impact of cardiac devices on the
quality of life in pediatric patients. Circulation: Arrhythmia and Electro-
physiology. 2012;5:1064–72.
13. Webster G, Panek KA, Labella M. Psychiatric functioning and qual-
ity of life in young patients with cardiac rhythm devices. Pediatrics.
2014;133:e964–e972.
14. Telford RM, Telford RD, Olive LS, Cochrane T, Davey R. Why are girls
less physically active than boys? Findings from the LOOK Longitudinal
Study. PLoS ONE. 2016;11:e0150041.
15. Faerestrand S, Breivik K, Ohm OJ. Assessment of the work capacity
and relationship between rate response and exercise tolerance asso-
ciated with activity-sensing rate-responsive ventricular pacing. Pacing
Clin Electrophysiol. 1987;10:1277–1290.
16. Adamson PB, Smith AL, Abraham WT. Continuous autonomic
assessment in patients with symptomatic heart failure prog-
nostic value of heart rate variability measured by an implanted
cardiac resynchronization device. Circulation. 2004;110:2389–
2394.
17. Caspersen CJ, Pereira MA, Curran KM. Changes in phys-
ical activity patterns in the United States, by sex and
cross-sectional age. Med Sci Sports Exercise. 2000;32:1601–
1609.
1290
18. Vincent SD, Pangrazi RP. An examination of the activity patterns of ele-
mentary school children. Pediatric Exercise Sci. 2002;14:432–441.
19. Lindquist CH, Reynolds KD, Goran MI. Sociocultural determinants of
physical activity among children. Prevent Med. 1999;29:305–312.
20. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M.
Physical activity in the United States measured by accelerometer. Med
Sci Sports Exercise. 2008;40:181.
DE LA UZ ET AL .
How to cite this article: de la Uz CM, Burch AE, Gunderson
B, Koehler J, Sears SF. How active are young cardiac device
patients? Objective assessment of activity in children with car-
diac devices. Pacing Clin Electrophysiol. 2017;40:1286–1290.
https://doi.org/10.1111/pace.13197