Europace (2015) 17, 787–793 CLINICAL RESEARCH

doi:10.1093/europace/euv003 Pacing and resynchronization therapy

Association between hospital procedure volume

and early complications after pacemaker
implantation: results from a large, unselected,
contemporary cohort of the German nationwide
obligatory external quality assurance programme
Bernd Nowak 1*, Karl Tasche 2, Linda Barnewold2, Günther Heller 2, Boris Schmidt 1,
Stefano Bordignon 1, K.R. Julian Chun 1, Alexander Fürnkranz 1, and Rajendra H. Mehta3
1
CCB, Cardioangiologisches Centrum Bethanien, Im Pruefling 23, Frankfurt a.M. D-60389, Germany; 2AQUA-Institute, Göttingen, Germany; and 3Duke Clinical Research Institute
at Duke University Medical Center, Durham, NC, USA

Received 10 December 2014; accepted after revision 5 January 2015; online publish-ahead-of-print 2 March 2015

Aims Several studies demonstrated an inverse relationship between cardioverter-defibrillator implantation volume and
complication rates, suggesting better outcomes for higher volume centres. However, the association of institutional
procedural volume with patient outcomes for permanent pacemaker (PPM) implantation remains less known, especially
in decentralized implantation systems.
.....................................................................................................................................................................................
Methods We performed retrospective examination of data on patients undergoing PPM from the German obligatory quality
and results assurance programme (2007–12) to evaluate the relationship of hospital PPM volume (categorized into quintiles of
their mean annual volume) with risk-adjusted in-hospital surgical complications (composite of pneumothorax, hae-
mothorax, pericardial effusion, or pocket haematoma, all requiring intervention, or device infection) and pacemaker
lead dislocation. Overall 430 416 PPM implantations were documented in 1226 hospitals. Systems included dual
(72.8%) and single (25.8%) chamber PPM and cardiac resynchronization therapy (CRT) devices (1.1%). Complications
included surgical (0.92%), and ventricular (0.99%), and atrial (1.22%) lead dislocation. Despite an increase in relatively
complex procedures (dual chamber, CRT), there was a significant decrease in the procedural and fluoroscopy times
and complications from lowest to highest implantation volume quintiles (P for trend ,0.0001). The greatest difference
was observed between the lowest (1–50 implantations/year-reference group) and the second-lowest (51–90 implanta-
tions/year) quintile: surgical complications [odds ratio (OR) 0.69; confidence interval (CI) 0.60– 0.78], atrial lead disloca-
tions (OR 0.69; CI 0.59– 0.80), and ventricular lead dislocations (OR 0.73; CI 0.63– 0.84).
.....................................................................................................................................................................................
Conclusions Hospital annual PPM volume was directly related to indication-based implantation of relatively more complex PPM
and yet inversely with procedural times and rates of early surgical complications and lead dislocations. Thus, our data
suggest better performance and lower complications with increasing procedural volume.
-----------------------------------------------------------------------------------------------------------------------------------------------------------
Keywords Pacemaker implantation complications † Lead dislocation † Implantation volume † Quality assurance

cardiovascular procedures suggesting better outcomes for higher

Introduction volume centres.1 – 3 A similar association between higher volume
Many studies have demonstrated an inverse relationship between and better outcomes has been shown for implantation of cardio-
institutional procedural volume and complications for diverse verter-defibrillators (ICDs).4,5 However, data on volume outcomes

* Corresponding author. Tel: +49 69 9450280; fax: +49 69 461613. E-mail address: b.nowak@ccb.de
Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: journals.permissions@oup.com.

Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065

by guest
on 17 November 2017
 788
What’s new?
† A very large, unselected, contemporary real-life population
and a wide range of hospitals were studied to examine the re-
lationship between pacemaker implantation volume and early
procedural complications.
† Hospital annual pacemaker implantation volume was directly
related to implantation of more complex pacemakers.
† Hospital annual pacemaker implantation volume was inversely
related to procedural times and cephalic vein access.
† Hospital annual pacemaker implantation volume was inversely
related to early surgical complications and early lead
dislocations.
relationship in contemporary clinical practice are scarce for pacemaker
implantations. Available data are derived from a centralized healthcare
system such as Denmark where only few centres perform pacemaker
implantation unlike in most other countries like Germany where this
procedure is performed in a wide variety of institutions, limiting its
generalizability.6
Therefore, the aim of this study is to evaluate the routinely
collected and available information in an unselected contemporary
large-volume real-life population from the German external obliga-
tory quality assurance programme to examine the relationship
between hospital pacemaker implantation volume and early proced-
ural complications requiring an intervention. We hypothesized
that, as observed for other cardiovascular procedures, an inverse
relationship existed between institutional pacemaker procedural
volume and post-implantation in-hospital complications.
Methods
German statutory external quality assurance
programme and the study population
and data collection
The German statutory external quality assurance programme—con-
ducted by the AQUA Institute (Institute for Applied Quality Improve-
ment and Research in Health Care, Göttingen, Germany) on behalf of
the Federal Joint Committee—collects real-life data on clinical and
procedural characteristics and in-hospital outcomes of all inpatient
primary permanent pacemaker (PPM) implantations in the country
across a wide range of hospitals varying in their procedural volumes.
The participation in this programme is linked to reimbursement of
the procedures, thus accounting for the data on over 95% of all PPM
implantations performed in Germany.7
For this study, the database of the statutory external quality assurance
programme in Germany was evaluated retrospectively for the years
2007 – 12. During these years, there was no change in the collected
dataset. Hospitals performing the procedure reported the data in
the programme prospectively on an ongoing basis. A total of 430 416
permanent primary pacemaker implantations were documented at
1226 hospitals in the database in this time frame.
For every in-hospital primary PPM implantation, a dataset has to be
filed electronically. These data are collected centrally for each federal
state in Germany on a yearly basis. Thereafter, all data are pooled in
one database, which allows a nationwide analysis of all implantations.
Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065
by guest
on 17 November 2017
B. Nowak et al.
On the basis of these data predefined quality benchmarks in terms of
outcomes have to be reached by each implanting centre. These are, for
example, achieving appropriate indication for pacemaker implantation
according to current guidelines in ≥90% of implantations. If those bench-
marks are not met, a ‘structured dialogue’ is initiated, to uncover the
reasons for the deviation and to trigger individual institution designed
quality improvement measures, if necessary, to improve outcomes.
Every hospital in Germany is required to publish quality assurance data.
The data collected represent information for the period from implant-
ation until discharge. The quality control programme collects overall
institutional data without differentiating between individual departments
or implanters.
In-hospital outcomes
Outcomes of interest for this study included (i) surgical complications—a
composite of pneumothorax, haemothorax, pericardial effusion, or
pocket haematoma, all requiring intervention, or device infection; and
(ii) atrial or ventricular lead dislocation.
Statistical analysis
For each implanting centre, the mean annual implantation volume was
computed by dividing the total number of cases in the study period
divided by the number of years the data was provided. Hospitals were
categorized into quintiles of their mean annual pacemaker implantation
volume. Continuous variables are reported as mean and standard
deviations; unordered categories as percentages; ordered categories
as median, 25th, and 75th percentiles. Univariate comparisons were
performed using the Pearson x 2 test for unordered categories, the
Kruskal – Wallis test for ordered categories, and one-way ANOVA
with polynominal contrasts for continuous variables (results for the
linear contrast are reported).
Multivariable logistic regression analyses were used to assess the
relationships between hospital volume and the outcomes of interest.
Adjustments were made for the following patient and procedural charac-
teristics: gender, age, American Society of Anesthesiologists (ASA) phys-
ical status, and type of device implanted [single-chamber, dual-chamber,
or cardiac resynchronization therapy pacemaker (CRT-P) systems].
The reference category was the quintile with lowest mean procedural
volume. Odds ratios (ORs) and 95% confidence intervals (CIs) were gen-
erated. For all analyses, a two-tailed P-value of ,0.05 was considered
statistically significant. All statistical analyses were performed using
SPSS 22.0 and Stata 11.1 for Windows.
Results
Study population
Overall 430 416 primary in-hospital pacemaker implantations
were documented at 1226 hospitals from 2007 to 2012 in
Germany. Of these, 656 hospitals were categorized in the lowest
volume quintile that implanted between 1 and 50 devices annually
(mean 23.67 + 14.20 implantations/year). The highest quintile con-
sisted of 55 hospitals that implanted more than 190 pacemakers
yearly (mean 280.88 + 143.98 implantations/year) (Table 1).
The mean age of 229,896 male patients (53.4%) was 74.5 + 10.5
and that of 200 520 female patients (46.6%) was 77.6 + 10.2. The
indication for pacemaker implantation included high-degree AV
block (40.3%), sick sinus syndrome (37.0%), atrial fibrillation with
bradycardia (19.2%), carotid sinus hypersensitivity syndrome and
vasovagal syncope (1.5%), and ‘other’ (3.1%) (Table 1).
 Pacemaker implantation volume and complications 789

Table 1 Study population, major pacemaker indications, and procedural parameters

Quintiles P-value
Implantations/year
......................................................................................................
Q1 Q2 Q3 Q4 Q5
1– 50 51– 90 91 –130 131– 190 >190
...............................................................................................................................................................................
Hospitals 656 259 152 104 55
Implantations 79 639 96 593 86 380 89 112 78 692
Implantations/year (mean + SD) 23.67 + 14.20 68.46 + 11.26 106.74 + 10.71 157.82 + 17.76 280.88 + 143.98
Age (mean + SD) (years) 77.4 + 9.5 76.9 + 9.3 76.2 + 9.6 74.9 + 11.4 74.2 + 12.1 ,0.0001a
Male (%) 50.3 52.6 53.1 55.3 55.7 ,0.0001b
Female (%) 49.7 47.4 46.9 44.7 44.3
ASA status
Median 3 2 2 2 2 ,0.0001c
First and third quartile 2-3 2-3 2-3 2-3 2-3
Mean rank 233.745,1 210.941,6 205.508,0 209.400,9 218.911,2
Indication AV block (%) 37.5 37.7 39.3 41.8 43.5
Indication atrial fibrillation with bradycardia (%) 22.7 21.4 18.4 17.1 14.9
Indication sick sinus syndrome (%) 35.1 36.8 38.3 36.5 35.9
Single-chamber systems (%) 31.6 28.4 24.2 23.8 20.9 ,0.0001b
Dual-chamber systems (%) 67.8 70.7 74.8 74.3 76.6 ,0.0001b
CRT-P systems (%) 0.33 0.70 0.91 1.59 1.95 ,0.0001b
Single-chamber implant duration (min) 46.9 + 24.8 42.5 + 20.9 41.5 + 21.3 41.2 + 27.6 38.8 + 30.0 ,0.0001a
Dual-chamber implant duration (min) 65.6 + 29.4 60.2 + 25.1 58.0 + 25.1 56.1 + 27.2 51.4 + 26.3 ,0.0001a
Single-chamber fluoroscopy (min) 3.7 + 4.3 3.4 + 3.9 3.3 + 4.1 3.2 + 4.1 3.2 + 4.3 ,0.0001a
Dual-chamber fluoroscopy (min) 5.9 + 5.5 5.4 + 4.9 5.3 + 4.9 5.1 + 4.9 4.8 + 4.8 ,0.0001a
Cephalic vein access (%) 43.9 39.6 37.5 34.0 28.2 ,0.0001a

a
Test for linear trend.
b 2
x test.
c
Kruskal –Wallis test.

Procedural characteristics
The following pacemaker systems were implanted: 313 374 dual- Pneumothorax Haematoma Pericardial effusion
chamber pacemakers (72.8%), 111 023 single-chamber pacemakers
0.8
(25.8%), and 4687 CRT-P systems (1.1%) [unspecified: 1332
.60 .58
(0.31%)]. The distribution of implanted system according to implant-
ation volume quintiles is shown in Table 1. With increasing implant- 0.6 .49
Incidence (%)

ation volume, the percentage of single-chamber systems decreased .42

.38
and the proportion of complex procedures (dual-chamber and 0.4 .34 .35

CRT-P system implantation) increased, most likely related to lower .26

incidence of atrial fibrillation and higher incidence of AV block as .21 .18
.17 .19
.16 .16
well as younger age at hospitals with increasing volumes. Increasing 0.2 .14

implantation volume was associated with shorter implantation

(skin-to-skin) and fluoroscopy times for both single- and dual- 0.0
chamber pacemakers and with lower use of the cephalic vein as com- 1–50 51–90 91–130 131–190 >190
pared with subclavian vein puncture (Table 1). For CRT-P systems, Implantations/year
which represented only 1.1% of implanted systems, time for implant-
ation was shortest in the highest quintile. Figure 1 Rates of the three most frequent surgical complications
according to implantation volume quintiles.

Complications
Overall in-hospital surgical complications occurred in 0.99% of device infection in 0.05%. Figure 1 shows the observed surgical com-
patients that included pneumothorax in 0.44%, pocket haematoma plication rates in various implant volume quintiles. Figure 2 depicts the
in 0.31%, pericardial effusion in 0.16%, haemothorax in 0.05%, and adjusted OR and 95% CI in the five volume categories. As shown, the

Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065

by guest
on 17 November 2017
 790 B. Nowak et al.

Surgical complications Ventricular lead dislocation

1–50 1.00 1–50 1.00

Implantations/year
Implantations/year

51–90 0.69 (0.60–0.78) 51–90 0.73 (0.63–0.84)

91–130 0.70 (0.60–0.80) 91–130 0.58 (0.50–0.68)

131–190 0.55 (0.48–0.64) 131–190 0.42 (0.35–0.50)

>190 0.64 (0.50–0.82) >190 0.39 (0.30–0.50)

0
0

0.

0.

1.
0.

0.

1.
OR (± 95% Cl) OR (± 95% Cl)

Figure 2 Odds ratio for surgical complications adjusted for
gender, age, ASA physical status, and pacing system according to
implantation volume quintiles (CI).
Figure 4 Association between pacemaker implantation volume
and ventricular lead dislocation, demonstrating the decrease in dis-
locations with increasing implantation volume. Odds ratio for ven-
tricular lead dislocation adjusted for gender, age, ASA physical
status, and pacing system according to implantation volume quin-
tiles (CI).

Atrial lead dislocation Ventriclar lead dislocation

2.5
2.11
2.0
Atrial lead dislocation

1.5 1.34
1.19 1–50 1.00
(%)

1.52
Implantations/year

1.0 0.79 0.80 51–90 0.69 (0.59–0.80)

1.07
0.85
0.5 91–130 0.61 (0.50–0.73)
0.61 0.57
131–190 0.40 (0.33–0.49)
1–50 51–90 91–130 131–190 >190
>190 0.39 (0.30–0.53)
Implantations/year
0

0
0.

0.

Figure 3 Observed atrial and ventricular lead dislocation rates
according to implantation volume quintile.
lowest volume category demonstrated the highest surgical complica-
tion rates with the greatest difference noted between the two lowest
volume hospital quintiles.
Ventricular and atrial lead dislocations occurred in 0.92 and 1.22%
of all implantations and decreased with increasing procedure volume
(Figure 3). Figures 4 and 5 display the adjusted ventricular and atrial
lead dislocation rates in various quintiles demonstrating significantly
lower odds of these leads displacements with increasing institutional
volume. Again, the maximum difference in lead displacement was
observed between the two lowest volume quintiles.
Gender-specific data for surgical complications and lead disloca-
tions are shown in Table 2. The overall rate of surgical complications
was higher in women than in men. Surgical complications and lead dis-
locations were highest among both men and women in the lowest
volume implantation quintile.
Discussion
Main findings
Our data in a large, unselected, contemporary national population
demonstrated that compared with hospitals with lower pacemaker
1.
OR (± 95% Cl)
Figure 5 Odds ratio for atrial lead dislocation adjusted for
gender, age, ASA physical status, and pacing system according to
implantation volume quintiles (CI).
implantation volume, those with higher volume were on an average
more likely to perform more complex implantations such as dual-
chamber pacemakers or to a low-percentage CRT-P devices
(rather than single-chamber pacemakers). Yet, despite increased im-
plantation of relatively more complex devices by higher volume
centres, performance increased as reflected by the lower procedural
and fluoroscopy times compared with the centres in the lowest
volume quintile. Most importantly, our data demonstrated an
inverse relationship between procedural volume and in-hospital
complications suggesting significantly better short-term outcomes
for higher volume hospitals compared with those in the lowest
volume quintile. This inverse relationship between procedural
volume and adverse events was not only restricted to early surgical
complications such as pneumothorax, haemothorax, pericardial effu-
sion, pocket site haematoma, or infection of device but also included
technical failures such as atrial and/or ventricular lead dislodgement.
Furthermore, even after risk adjustment for baseline confounders,
higher volume centres continued to have lower complications and

Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065

by guest
on 17 November 2017
 Pacemaker implantation volume and complications
Table 2 Gender-specific complications
Quintiles implantations/year
.................................................................................................................
Q1 Q2
1–50 51 –90
...............................................................................................................................................................................
Surgical complications
Female (%) 1.83a 1.19
Male (%) 1.05a 0.73
Atrial lead dislocation
Female (%) 1.88a 1.15
Male (%) 2.33a 1.51
Ventricular lead dislocation
Female (%) 1.62a 1.18
Male (%) 1.43a 0.98
a
Test for linear trend.
lead dislodgement compared with those in the lowest volume quin-
tile. Thus, higher institutional pacemaker volumes were associated
with better performance and outcomes despite higher use of sub-
clavian venous access and increasing number of complex device
implantation-characteristics associated and higher complication
rates based on previous investigations.6 – 10
Finally, our data showed that the overall rate of any surgical com-
plications was higher in women than in men, both sexes were less
likely to have adverse events or lead dislodgement at higher
volume sites compared with those in the lowest volume quintile—
a trend similar to that observed in the overall population.
Comparison with prior studies
Two large studies have examined the relationship of institutional
annual ICD procedural volume and outcomes and reported
inverse relationship between procedural volume and complications
that were evident across all types of ICD subtypes, i.e. single-
chamber, dual-chamber or biventricular ICD devices.4,5 A European
survey on CRT implantation found significantly shorter implantation
and fluoroscopy times and lower rates of pocket haematoma for
high-volume centres, but no difference in other complications com-
pared with lower volume hospitals.11 Recently, published data from
Denmark evaluated 6-month complications in 5918 consecutive
cardiac implantable electronic device performed from 2010 to
2011, including pacemakers, ICDs, and CRT systems.6 They demon-
strated that the risk of complications was higher in female patients,
underweight patients, at centres with an annual volume ,750 proce-
dures or by operators with ,50 procedures, with dual-chamber
ICD, for procedures involving system upgrade or lead revision, and
in those who underwent an emergency off-hour procedure. The
same group of investigators in their earlier study from the Danish
Pacemaker Register of 28 860 pacemaker implantations between
1997 and 2008, showed a lead complication rate of 2.3% for atrial
and of 2.2% for right ventricular leads at 3 months follow-up with
increased rates among inexperienced operators with ,25 implanta-
tions and with dual-chamber system implantation.12 In contrast, a
study from the Netherlands evaluating 1517 implantations over a
Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065
by guest
on 17 November 2017
791
P-valuea
Q3 Q4 Q5
91 –130 131– 190 >190
1.28 1.0 1.17 ,0.0001
0.65 0.57 0.65 ,0.0001
1.13 0.65 0.74 ,0.0001
1.25 0.91 0.84 ,0.0001
0.92 0.71 0.69 ,0.0001
0.78 0.52 0.46 ,0.0001
4.5-year period found that annual hospital implanting volume did
not contribute to the prediction of short-term complications.13
A small study from Norway compared 535 pacemaker implantations
in two hospitals with annual implantation rates of 28 and 84 (volumes
that were similar to that in the lowest 2 quintiles in our study).14 The
overall complication rate was high at 12%, major complications
occurred in 7.5%, and reoperation was required in 5.2%. There was
no statistically significant difference between the two hospitals in out-
comes. In this study, procedures performed by trainees (marker of
lower volume and inexperience) were associated with higher rates
of complications. Further support to the relationship of higher
volume (greater experience) with better outcomes comes from
another small study. Pakarinen et al.15 evaluated outcomes of 567
cardiac rhythm management devices implantations, including
upgrades and revisions, at a tertiary referral university hospital in
Finland and found that the complication rates were more than
two-fold higher if pacemaker implantations were performed by car-
diology trainees compared with experienced cardiologists (17.4% vs.
7.7%, P ¼ 0.001).
Most of these studies along with our findings support a direct
relationship of procedural volumes with outcomes after cardiac
rhythm device implantations. However, differences between these
and our study need to be highlighted. First, unlike some of the
above studies that focused on ICD or CRT, our study focused only
on pacemaker implantations. Secondly, the study from the Danish
Pacemaker Register highlights fundamental differences between the
situation in Denmark and in Germany. In Denmark, device implanta-
tions are centralized to 14 hospitals unlike in Germany where they
are performed in more than 1000 hospitals. Therefore, the lowest
volume centres in Denmark perform ,249 procedures (pacemaker
and ICD), and low-volume implanters perform ,50 implantations
per year.6 In contrast, in Germany more than 650 hospitals have
an annual pacemaker implantation volume ,50 and only 55 have
an implantation volume .190, thus allowing us to examine the
volume–outcomes relationship across a wide range of procedural
volumes in contemporary practice. Unlike some of the above
studies, we only reported in-hospital complications.
 792
Prior studies have also demonstrated that subclavian vein access is
associated with an increased risk of complications, especially
pneumothorax.7,9,10,16 – 19 In the Danish population-based study of
28 860 patients, the risk factors associated with increased risk of
pneumothorax were female gender, age .80, chronic obstructive
pulmonary disease, dual-chamber pacemaker implantation, sub-
clavian vein puncture, and implantation in a non-university centre.16
That the rate of pneumothorax did not significantly differ among
the hospitals in the different volume categories may be a reflection
of the increasing rate of subclavian puncture (as oppose to cephalic
vein preparation) with higher implantation volume. It remains to be
proven whether increasing the use of cephalic vein access in high-
volume centres may lead to an even greater difference in the rates
of pneumothorax between low- and high-volume centres and to
what extent will it increase procedural time.
Finally, our study results were consistent with those of previous
studies that demonstrated that women have higher procedural com-
plication rate than men with device implantation.6,17,19 However, our
study further showed that both sexes were more likely to have better
outcomes in higher volume centres compared with those in the
lowest volume quintile.
Implications of our study findings
Data from previous studies and our findings suggests that the inverse
volume–outcomes relationship is nearly continuous even for the
relatively simpler procedure of PPM implantation (compared with
CRT or ICD implantation). Furthermore, perhaps worse outcomes
are observed for institutions implanting ,50 pacemaker implants
per year. It has to be discussed, whether it might be best for these hos-
pitals to defer pacemaker implantations to centres with higher
volumes and potentially by high-volume operators to decrease com-
plications and improve outcomes. This may be particularly reason-
able when implantation of advanced devices such as dual-chamber
pacemaker, CRT-P, ICDs, or biventricular ICD is required. In add-
ition, patients at higher risk of complications after pacemaker
implantation, i.e. elderly, women, underweight, those with chronic
obstructive pulmonary disease, those requiring lead revision, may
also potentially benefit from implantation at higher volume sites.
On the other hand, restricting pacemaker implantations to fewer
high-volume hospitals has the potential for overwhelming these
centres and likely to result in longer wait for the procedure, which
may have the potential for an adverse event while waiting as shown
by a previous study.20
Limitations
The data collection of the quality assurance programme ends with
discharge of the patient from the implanting hospital. Mandatory
submission of data on longer follow-up will be implemented in the
near future, but is not realized yet. Therefore, we are unable to
provide information on relationship of institutional pacemaker
procedural volume with long-term outcomes. For this reason, the
rate of all complications (including device infections) was very low
in our study compared with previous reports. Only those surgical
complications that required specific interventions were reported.
Thus, we do not have data on complications, such as pneumothorax,
haemothorax, pericardial effusion, or haematoma, managed conser-
vatively. Therefore, our data are potentially confounded by the
Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065
by guest
on 17 November 2017
B. Nowak et al.
aggressiveness of complication management in the individual
centre and reporting bias. Only inpatient pacemaker implantations
are evaluated by the quality assurance programme. Outpatient
procedures are not yet included. We were only able to adjust for
measured confounders—a limitation inherent with any registry ana-
lysis of already collected data. Data collected were centre-specific
rather than operator-specific precluding examination of operator
volume or specialty specific (cardiologists, electrophysiologists, and
surgeons) relationship with outcomes. While on an average we
found better outcomes for higher volume centres, there was signifi-
cant variability among these groups and individual sites in different
volume categories. Thus, some centres in the lowest quintile may
have had better outcomes than some higher volume institutions.
Although the data collection for quality assurance purposes is obliga-
tory and linked to operation and procedure coding and diagnosis-
related group classification, there is no formal ongoing audit to
validate the data. Nevertheless, a data validation programme
performed in 2008 compared patient files with the documented
quality assurance data and showed data correctness .99% in the
parameters evaluated in this study.
Conclusions
Hospital annual pacemaker implantation volume was directly related
to indication-based implantation of more complex pacemakers and
yet inversely with procedural times and rates of surgical complica-
tions and lead dislocations. Thus, our data suggest better perform-
ance and lower complications with increasing procedural volume.
Future studies should evaluate the risks vs. benefits of longer wait
to referral for pacemaker implantation to higher volume centres
compared with implantation of these at lower volume hospitals.
Conflict of interest: B.N. has received speaker fees from Biotronik.
B.S. is an advisory board member and has received speaker fees from
St Jude Medical and Boston Scientific. S.B. has received speaker fees
from Medtronic. K.R.J.C. is an advisory board member and has
received speaker fees from Medtronic. A.F. has received speaker
fees from Medtronic. R.M. has received research grants through
Duke Clinical Research Institute from St Jude Medical and Tenax
Therapeutics, Inc.
References
1. Post PN, Kuijpers M, Ebels T, Zijlstra F. The relation between volume and outcome
of coronary interventions: a systematic review and meta-analysis. Eur Heart J 2010;
31:1985 –92.
2. Wijns W, Kolh PH. Experience with revascularization procedures does matter: low
volume means worse outcome. Eur Heart J 2010;31:1954 –7.
3. West RM, Cattle BA, Bouyssie M, Squire I, de Belder M, Fox KAA et al. Impact of
hospital proportion and volume on primary percutaneous coronary intervention
performance in England and Wales. Eur Heart J 2011;32:706 –11.
4. Freeman JV, Wang Y, Curtis JP, Heidenreich PA, Hlatky MA. The relation between
hospital procedure volume and complications of cardioverter-defibrillator implant-
ation from the implantable cardioverter-defibrillator registry. J Am Coll Cardiol 2010;
56:1133 –9.
5. Al-Khatib SM, Lucas FL, Jollis JG, Malenka DJ, Wennberg DE. The relation between
patients’ outcomes and the volume of cardioverter-defibrillator implantation proce-
dures performed by physicians treating medicare beneficiaries. J Am Coll Cardiol 2005;
46:1536 –40.
6. Kirkfeldt RE, Johansen JB, Nohr EA, Jorgensen OD, Nielsen JC. Complications
after cardiac implantable electronic device implantations: an analysis of a complete,
nationwide cohort in Denmark. Eur Heart J 2014;35:1186 – 94.
 Pacemaker implantation volume and complications
7. Nowak B, Misselwitz B. on behalf of the Expert Committee ‘Pacemaker’, Institute of
Quality Assurance Hessen. Effects of increasing age onto procedural parameters in
pacemaker implantation: results of an obligatory external quality control program.
Europace 2008;11:75 –9.
8. Eberhardt F. Long term complications in single and dual chamber pacing are influ-
enced by surgical experience and patient morbidity. Heart 2005;91:500 – 6.
9. van Eck JWM, van Hemel NM, Zuithof P, van Asseldonk JPM, Voskuil TLHM,
Grobbee DE et al. Incidence and predictors of in-hospital events after first implant-
ation of pacemakers. Europace 2007;9:884 – 9.
10. Chauhan A, Grace AA, Newell SA, Stone DL, Shapiro LM, Schofield PM et al. Early
complications after dual chamber versus single chamber pacemaker implantation.
Pacing Clin Electrophysiol 1994;17:2012 –5.
11. Bogale N, Priori S, Gitt A, Alings M, Linde C, Dickstein K et al. The European cardiac
resynchronization therapy survey: patient selection and implantation practice vary
according to centre volume. Europace 2011;13:1445 –53.
12. Kirkfeldt RE, Johansen JB, Nohr EA, Moller M, Arnsbo P, Nielsen JC. Risk factors for
lead complications in cardiac pacing: a population-based cohort study of 28,860
Danish patients. Heart Rhythm 2011;8:1622 –8.
13. Udo EO, Zuithoff NPA, van Hemel NM, de Cock CC, Hendriks T, Doevendans PA
et al. Incidence and predictors of short- and long-term complications in pacemaker
therapy: the FOLLOWPACE study. Heart Rhythm 2012;9:728 – 35.
Downloaded from https://academic.oup.com/europace/article-abstract/17/5/787/2467065
by guest
on 17 November 2017
793
14. Haug B, Kjelsberg K, Lappegard KT. Pacemaker implantation in small hospitals:
complication rates comparable to larger centres. Europace 2011;13:1580 –6.
15. Pakarinen S, Oikarinen L, Toivonen L. Short-term implantation-related complica-
tions of cardiac rhythm management device therapy: a retrospective single-centre
1-year survey. Europace 2010;12:103–8.
16. Kirkfeldt RE, Johansen JB, Nohr EA, Moller M, Arnsbo P, Nielsen JC. Pneumothorax
in cardiac pacing: a population-based cohort study of 28 860 Danish patients.
Europace 2012;14:1132 –8.
17. Nowak B, Misselwitz B, on behalf of the expert committee ‘Pacemaker’, Institute of
Quality Assurance Hessen, Erdogan A, Funck R, Irnich W et al. Do gender differences
exist in pacemaker implantation?—results of an obligatory external quality control
program. Europace 2010;12:210–5.
18. Link M, Estes N 3rd, Griffin J, Wang P, Maloney J, Kirchhoffer J et al. Complications
of dual chamber pacemaker implantation in the elderly. Pacemaker Selection
in the Elderly (PASE) Investigators. J Interv Card Electrophysiol. 1998;2:175 –9.
19. Peterson PN, Daugherty SL, Wang Y, Vidaillet HJ, Heidenreich PA, Curtis JP et al.
Gender differences in procedure-related adverse events in patients receiving
implantable cardioverter-defibrillator therapy. Circulation 2009;119:1078 –84.
20. Risgaard B, Elming H, Jensen GV, Johansen JB, Toft JC. Waiting for a pacemaker: is it
dangerous? Europace 2012;14:975–80.