Trends of Cardiovascular Implantable Electronic Device Infection in 3 Decades

JACC: CLINICAL ELECTROPHYSIOLOGY

VOL. 5, NO. 9, 2019

ª 2019 PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

Trends of Cardiovascular Implantable Electronic Device Infection in 3 Decades A Population-Based Study Mingyan Dai, MD,a,b Cheng Cai, MD,b,c Vaidya Vaibhav, MBBS,b M. Rizwan Sohail, MD,b,d David L. Hayes, MD,b David O. Hodge, MS,e Ying Tian, MD, PHD,b,f Roshini Asirvatham,b Jordan J. Cochuyt,e Congxin Huang, MD, PHD,a Paul A. Friedman, MD,b Yong-Mei Cha, MDb

ABSTRACT OBJECTIVES This study assessed trends in the incidence of cardiovascular implantable electronic device (CIED) infection in the last 3 decades using a population-based records linkage study. BACKGROUND Infection remains an important issue associated with increased implantation rate and dwell time of CIEDs. METHODS We identified a cohort of all adults with CIEDs who resided in Olmsted County, Minnesota, from 1988 to 2015, using the medical linkage system of the Rochester Epidemiology Project. Standardized criteria were used to identify all CIED infection cases. The cumulative rate of CIED infection was estimated using the Kaplan-Meier method, and the trends of CIED infection incidence were calculated with person-years of follow-up after device implantation. RESULTS The cumulative probabilities of overall CIED infection were 6.2% (95% confidence interval [CI]: 4.0% to 8.4%) at 15 years and 11.7% (95% CI: 6.8% to 17.3%) at 25 years of follow-up. The incidence of CIED infection every 7 years from 1988 to 2015 was 1.3, 5.7, 4.1, and 4.7 per 1,000-person years, respectively. The 15-year cumulative probabilities of CIED infection after the initial, second, and third procedures were 2.6% (95% CI: 1.4% to 3.8%), 2.7% (95% CI: 1.2% to 4.2%), and 24.1% (95% CI: 3.8% to 44.4%), respectively. Generator changes (hazard ratio [HR]: 3.91; 95% CI: 1.47 to 10.37; p ¼ 0.006) and upgrades (HR: 3.08; 95% CI: 1.24 to 7.62; p ¼ 0.02) were significantly associated with infection. CONCLUSIONS The incidence of CIED infection had a trend of increasing in the past 2 decades. Contemporary implantable cardioverter-defibrillator and cardiac resynchronization therapies and repeated manipulation of device pockets introduced a greater risk of CIED infection. (J Am Coll Cardiol EP 2019;5:1071–80) © 2019 Published by Elsevier on behalf of the American College of Cardiology Foundation.

T

he estimated incidence of cardiovascular

of proportion to that of CIED implantations (1–4). It

implantable electronic device (CIED) infec-

has been suggested that, due to an aging population,

tion increased from the 1990s to 2000s in

increases in CIED implantation volumes for expanded

the United States, and the rate of increase was out

indications and increases in multiple procedures after

From the aDepartment of Cardiology, Renmin Hospital of Wuhan University, Cardiovascular Research Institute, Wuhan University, Hubei Key Laboratory of Cardiology, Wuhan, Hubei, China; bDepartment of Cardiovascular Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota; cDepartment of Cardiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China; dDepartment of Internal Medicine, Division of Infectious Disease, Mayo Clinic College of Medicine and Science, Rochester, Minnesota;

e

Department of Health Sciences Research, Mayo Clinic College of Medicine and Science,

Rochester, Minnesota; and the fDepartment of Cardiology, Beijing Chaoyang Hospital, Beijing, China. This work was funded by the Department of Cardiovascular Diseases, Mayo Clinic. Dr. Sohail has been a consultant for Medtronic and Aziyo Biologics. Dr. Friedman has been a consultant for Medtronic and Boston Scientific; and has received research support from St. Jude Medical and Abbott. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page. Manuscript received March 1, 2019; revised manuscript received June 13, 2019, accepted June 27, 2019.

ISSN 2405-500X/$36.00

https://doi.org/10.1016/j.jacep.2019.06.016

1072

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

ABBREVIATIONS

primary implantation, the prevalence of

with CIEDs implanted elsewhere, who then relocated

AND ACRONYMS

CIED-related infection may increase in the

to Olmsted County, were also included if they had

future (3,5–8). Advances in our understand-

ever received medical care related to their device.

ing of CIED-related infection have improved

Patients with subsequent procedures at least 1 day

and standardized the care of patients with

after initial implantation were identified using the

CI = confidence interval CIED = cardiovascular implantable electronic devices

CIED infection, as well as improving prophy-

procedure codes (Online Table 2), and the charts of

lactic measures to prevent infections (9).

these patients at the Mayo Clinic were manually

However, whether these measures have

reviewed (M.D.) to ascertain procedure dates and

overcome the increasing risks for CIED infec-

types. Subsequent procedures were divided into 3

cardioverter-defibrillator

tion in recent years has not been well

main categories: 1) revisions that included lead re-

ICD-9-CM = International

studied.

visions, insertions, replacements, and pocket re-

CRT = cardiac resynchronization therapy

HR = hazard ratio ICD = implantable

Estimates of CIED infection rates in the

visions; 2) generator changes; and 3) upgrades, which

Revision, Clinical Modification

United States are mostly based on adminis-

included upgrading from a permanent pacemaker

PPM = permanent pacemaker

trative

databases.

(PPM) to an implantable cardioverter-defibrillator

REP = Rochester Epidemiology

Reliance on administrative claimsbased

(ICD) or cardiac resynchronization therapy (CRT)

databases has been shown to have low spec-

and upgrading from ICD to CRT. The first subsequent

ificity in identifying true CIED infection cases (10).

procedures after initial implantations were defined as

Therefore, it is imperative to assess the true incidence

the second procedures.

Classification of Diseases, Ninth

Project

and

hospital-based

of CIED infection over time using population-based studies. We conducted a large population-based study in Olmsted County, Minnesota, to investigate historical trends for CIED infection and to evaluate the relationship between subsequent procedures after initial implantation and CIED infection. SEE PAGE 1081

CIED INFECTION CASE DEFINITIONS. Patients with a

CIED infection, either a generator pocket infection or systemic infection including device lead-associated endocarditis, were identified in 1 of 2 ways (1,13): 1) a patient was diagnosed with device infection using the ICD-9-CM code (Online Table 3); or 2) a CIED removal code or lead extraction code, together with

METHODS

evidence of systemic infection, including sepsis,

DATA SOURCE. The Rochester Epidemiology Project

Table 3). The charts of CIED infection cases identi-

bacteremia, or fever during the same visit (Online

(REP) is a medical record system that includes medi-

fied by ICD-9-CM codes were then manually reviewed

cal records from all individuals residing in Olmsted

to ascertain the presence of CIED infection.

County. Each patient has a single dossier into which

CIED infection was confirmed by either microbio-

medical records, including medical diagnoses, surgi-

logical or clinical criteria (12–16). Microbiological de-

cal procedures, and other key information, are regu-

vice infection was confirmed based on positive

larly entered using the International Classification of

cultures from the generator pocket, leads, or blood (in

Diseases, Adapted Code for Hospitals (11,12). Because

the presence of local inflammatory signs at the

the REP provides access to details of health care of

generator pocket or the absence of another source of

local residents regardless of health provider, collect-

bacteremia and resolution of blood stream infection

ing accurate incidence data and population-based

after device explantation) (17).

analytical studies of diseases are possible (11). REP

STATISTICAL ANALYSIS. The cumulative rates of

dossiers include medical histories and diagnoses for

CIED infection and mortality after implantation were

all patient encounters within the health care system,

estimated using the Kaplan-Meier method. Patients

including both hospitals (Mayo Clinic and Olmsted

were censored at their last visit to the clinic. Univar-

Medical Center, Rochester). The number of clinicians

iate and multivariate analyses in a Cox proportional

who implanted CIED was 4 in early 1990s and 10 in

hazards model were used to assess patient risk factors

2015. Standardized implanting techniques, proper

of CIED infection. The multivariate model included

sterilization, and prophylactic intravenous antibiotics

univariate variables with p < 0.10. The cumulative

were used for each implantation.

incidence rate of CIED infection was calculated using

STUDY POPULATION. Patients were selected if they

incident cases of CIED infection as the numerator and

had a documented record of an initial CIED implan-

person-years of device implantation as the denomi-

tation from 1988 to 2015 using the International

nator. A Cox proportional hazards model with a

Classification

Revision-Clinical

time-dependent covariate was used to assess the as-

Modification (ICD-9-CM) code. The ICD-9-CM pro-

sociation of mortality with CIED infection and to

cedure codes are provided in Online Table 1. Patients

compare the infection rates of revisions, generator

of

Diseases-9th

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

changes, and upgrades. A p value < 0.05 was considered statistically significant. All analyses were

T A B L E 1 Baseline Patient Characteristics

All (N ¼ 2163)

performed on SAS version 9.4 (SAS Institute, Cary, North Carolina).

RESULTS

1073

Trends in CIED Infection

Age, yrs Male

CIED Infection (n ¼ 59)

No CIED Infection (n ¼ 2,104)

74  14

69  16

74  14

1,244 (57.5)

35 (59.3)

1,209 (57.5)

Comorbidities Hypertension

1,675 (77.4)

38 (64.4)

1,637 (77.8)

INCIDENCE OF CIED INFECTION. A total of 2,163

Hyperlipemia

1,417 (65.5)

39 (66.1)

1,378 (65.5)

patients with CIEDs were identified in Olmsted

Congestive heart failure

1,081 (50.0)

30 (50.8)

1,051 (50.0)

County (approximately 1.45% of 148,736 residents

Cardiomyopathy

607 (28.1)

17 (28.8)

590 (28.0)

Atrial fibrillation

1,083 (50.1)

31 (52.5)

1,052 (50.0)

Diabetes mellitus

1,832 (84.7)

51 (86.4)

1,781 (84.6)

Coronary artery disease

1,345 (62.2)

37 (62.7)

1,308 (62.2)

Myocardial infarction

751 (34.7)

19 (32.2)

732 (34.8)

Percutaneous transluminal coronary angioplasty

343 (15.9)

9 (15.3)

334 (15.9)

implantations gradually increased to 8.5%, 17.5%,

Perivascular disease

848 (39.2)

21 (35.6)

827 (39.3)

25.3%, and 28.6%, respectively (p < 0.001). In a

Cerebrovascular disease

739 (34.2)

17 (28.8)

722 (34.3)

Previous stroke

564 (26.1)

13 (22.0)

551 (26.2)

Transient ischemic attack

373 (17.2)

9 (15.3)

364 (17.3)

according to the American Community Survey 2015 estimate) between 1988 and 2015. The baseline demographics with and without infections are shown in Table 1. In 1988 to 1994, 1995 to 2001, 2002 to 2008, and 2009 to 2015, the proportions of ICD and CRT

28-year span, 62 infection cases occurred in 59 patients, among whom 3 patients had recurrent

Valvular heart disease

201 (9.3)

11 (18.6)

190 (9.0)

infections after device system removals and re-

Chronic kidney disease

573 (26.5)

14 (23.7)

559 (26.6)

implantation. The total number of follow-up per-

Chronic obstructive pulmonary disease

582 (26.9)

15 (25.4)

567 (26.9)

son-years from device implantation to infection,

Malignancy

879 (40.6)

20 (33.9)

859 (40.8)

death, or heart transplantation was 13,977 years. The

Hemodialysis dependency

41 (1.9)

0 (0.0)

41 (1.9)

Obesity (BMI >25 kg/m2)

1,032 (47.7)

27 (45.8)

1,005 (47.8)

incidence of CIED infection was 4.2 per 1,000 personyears (95% confidence interval [CI]: 1.2 to 7.9), including 2.9 for PPMs (95% CI: 0.3 to 4.6), 8.4 for ICDs (95% CI: 3.5 to 11.6), and 11.0 for CRTs (95% CI: 2.7 to 18.2). The incidence of infection after the first, second, and third procedure was 1.5% (33 of 2,163

Heart/liver/kidney transplant Sepsis Endocarditis

100 (4.6)

5 (8.5)

95 (4.5)

44.2  18.7

51.0  16.8

LVEDD (n ¼ 750)

52.6  9.0

54.4  8.9

52.5  9.0

Device type Single-chamber pacemaker Dual-chamber pacemaker

were bloodstream infections, including 7 cases of

42 (1.9) 183 (8.7)

50.8  16.8

patients). infections, including 9 with bacteremia; 21 (33.9%)

0 (0) 4 (6.8)

LVEF (%) (n ¼ 884)

patients), 2.9% (18 of 620 patients), and 5% (11 of 220 Of the 62 infection cases, 41 (66.1%) were pocket

42 (1.9) 187 (8.6)

427 (19.7)

7 (11.9)

420 (20.0)

1,254 (58.0)

24 (40.7)

1,230 (58.5)

ICD

373 (17.2)

22 (37.3)

351 (16.7)

CRT-P

20 (0.9)

1 (1.7)

19 (0.9)

CRT-D

89 (4.1)

5 (8.5)

84 (4.0)

endocarditis. Fifty-four (87%) cases had microbio-

Values are mean  SD or n (%).

logical diagnoses (Table 2). The cumulative probabil-

BMI ¼ body mass index; CIED ¼ cardiovascular implantable electronic device; CRT-D ¼ cardiac resynchronization therapy defibrillators; CRT-P ¼ cardiac resynchronization therapy pacemakers; ICD ¼ implantable cardioverter-defibrillator; LVEDD ¼ left ventricular end-diastolic dimension; LVEF ¼ left ventricular ejection fraction.

ities of CIED pocket or blood stream infection are shown in Figure 1A. The cumulative probability of pocket infection rose after 18 years, whereas blood stream infection appeared to be stable. Infection rates were significantly higher in the ICD and CRT groups

significant differences throughout the years in all age

than rates in the PPM group (CRT vs. PPM; p ¼ 0.002;

groups (p > 0.05 for all). The 50- to 70-year-old group

ICD vs. PPM; p < 0.001) (Figure 1B).

showed a greater increase in 2009 to 2015 compared

TRENDS OF CIED INFECTION. Trends of CIED infec-

tion in Olmsted County between 1988 and 2015 are shown in the Central Illustration. The incidence of CIED infection had an increasing trend (p ¼ 0.07) throughout the 28 years, regardless of device type. There was an apparent change in the incidence of infection from 1988 to 1994 to 1995 to 2001 (2.5- and 1.9-fold for the PPM and ICD groups, respectively).

with the other 2 age groups (p < 0.02). Within this group, 83% had ICD or CRT infections, which was significantly higher than that in the 70 years or older group (14%; p ¼ 0.03). Only 1 infection occurred in the older than 50 years group from 2009 to 2015. There was no significant difference in CIED infection between men and women. INFECTION IN INITIAL AND SUBSEQUENT CIED

The trends of CIED infection in age and sex groups

PROCEDURES. Of the 62 infection cases, 33 (53.2%)

are shown in Figures 2A and 2B. There were no

occurred after initial implantation, 18 (29.0%) after

1074

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

1.5%; 95% CI: 0.3% to 2.7%; p ¼ 0.02). After the third

T A B L E 2 Microbiologic Distribution of CIED Infections

procedure, there was a significant difference among Infections

the PPM, ICD, and CRT groups (p ¼ 0.002), and the

Staphylococcus aureus

27 (43.5)

15-year infection rate was significantly higher in the

Coagulase-negative staphylococcus

16 (25.8)

ICD group than that in the PPM group (44.6%; 95% CI:

Enterococcus faecalis

8 (12.9)

Cutibacterium/Propionibacterium acnes

4 (6.5)

6.6% to 82.5% vs. 5.4%; 95% CI: 0% to 11.8%; p ¼ 0.03).

Others*

4 (6.5)

Organism

The 10-year cumulative probabilities of infection after lead revision, generator change, and upgrade

Values are n (%), *Includes Escherichia coli, Stenotrophomonas maltophilia, Viridans group of streptococci, and Cutibacterium/Propionibacterium avidum.

were 2.8% (95% CI: 0.4% to 5.2%), 7.1% (95% CI: 0.7% to 13.3%), and 6% (95% CI: 0% to 14%), respectively (Online Figure 1). Both generator changes and upgrades had significant associations with infection

the second procedure, and 11 (17.7%) after the third and more procedures. The median (interquartile range) time from the last CIED procedure to the infections was 0.6 years (0.1 to 3.2 years). The 15-year

compared with initial implantations (hazard ratio [HR]: 3.08; 95% CI: 1.24 to 7.62; p ¼ 0.02; HR: 3.91; 95% CI: 1.47 to 10.37; p ¼ 0.006, respectively) (Figure 4).

cumulative probabilities of CIED infection after the initial, second, and third procedures were 2.6%

PREDICTORS OF INFECTION AND SURVIVAL. When

(95% CI: 1.4% to 3.8%), 2.7% (95% CI: 1.2% to 4.2%),

individual baseline characteristics were assessed,

and 24.1% (95% CI: 3.8% to 44.4%), respectively

diabetes mellitus, valvular heart disease, ICD, and

(Figure 3). There was no significant difference in

CRT were independent predictors of infection by a

infection rates between different types of CIEDs after

multivariate model (Table 3). In a median (inter-

initial implantations (PPM: 2.6%; 95% CI: 1.2% to

quartile range) follow-up of 5.8 years (3.0 to 10.3

4.0%; ICD: 2.0%; 95% CI: 0.5% to 3.5%; CRT: 4.3%;

years), 1,281 patients died. Using infection as a time-

95% CI: 0% to 8.6%; p $0.14). After the second pro-

dependent covariate of mortality in a Cox propor-

cedure, there were no infections in the CRT group,

tional hazards model, there was no significant

and the 15-year cumulative probability of infection

difference in mortality between infection and non-

was significantly higher in the ICD group compared

infection groups (HR: 0.87; 95% CI: 0.22 to 3.50;

with the PPM group (6.3%; 95% CI: 1.5% to 17.4% vs.

p ¼ 0.85).

F I G U R E 1 Cumulative Probability of Infection in Patients

(A) Cumulative probability of infection in patients with all cardiovascular implantable electronic devices (CIEDs) and 2 specific syndromes and (B) in the permanent pacemaker (PPM), implantable cardioverter-defibrillator (ICD), and cardiac resynchronization therapy (CRT) groups in a follow-up of 25 years. The cumulative probabilities of overall CIED infection were 6.2% (95% confidence interval [CI:] 4.0% to 8.4%) at 15 years and 11.7% (95% CI: 6.8% to 17.3%) at 25 years. The 15-year cumulative probabilities in the PPM, ICD, and CRT groups were 4.1% (95% CI: 2.1% to 6.0%), 12.1 (95% CI: 0% to 35.9%), and 22.8% (95% CI: 0% to 51.1%), respectively.

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

C ENTR AL I LL U STRA T I O N Incidence of CIED Infection

Probability of Infection

0.4

0.3

0.2

0.1

0.0 0.0

1988-1994 1995-2001 2002-2008 2009-2015

282 512 708 661

2.5

237 448 625 604

213 404 574 557

192 365 532 453

5.0

173 333 485 341

160 299 444 230

7.5 10.0 Infection (Time)

12.5

15.0

138 124 111 101 92 82 70 63 55 50 271 237 205 185 160 146 128 120 111 94 396 340 308 274 211 160 105 67 42 19 0 7 147 89 45

Dai, M. et al. J Am Coll Cardiol EP. 2019;5(9):1071–80.

Incidence of CIED infection every 7 years from 1988 to 2015 in the overall CIED group. Abbreviations as in Figure 1.

F I G U R E 2 Incidence in Different Age and Sex Groups

(A) Incidence of CIED infection every 7 years from 1988 to 2015 in different age and (B) sex groups. Abbreviations as in Figure 1.

1075

1076

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019

Trends in CIED Infection

SEPTEMBER 2019:1071–80

F I G U R E 3 Cumulative Probabilities of CIED Infection After the Initial, Second, and Third Procedures

Cumulative probability of CIED infection after the (A) initial implantation, (B) the second procedure, and (C) third procedure. Abbreviations as in Figure 1.

1077

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

F I G U R E 4 Association of Different Procedure Types

Association of different procedure types after initial implantation with CIED infection. The initial implantation was used as reference. CI ¼ confidence interval; HR ¼ hazard ratio; other abbreviation as in Figure 1.

DISCUSSION

There was no significant difference in infection incidence

observed

between

the

sex

groups

KEY FINDINGS. Our study found: 1) the incidence of

throughout the years. The 2017 Heart Rhythm Society

infection had an increasing trend since the late 1990s;

consensus statement on CIED lead management

2) more than one-half of CIED infection events

indicated that older age, along with increased

occurred after the de novo implantation; 3) the cumulative probability of infection was significantly higher in ICD and CRT groups than that in the PPM group; and 4) the CIED infection rate increased sub-

T A B L E 3 Patient Risk Factors for CIED Infection

Univariate Model

stantially after the third procedure compared with the first and second procedures. Age

Multivariate Model

HR (95% CI)

p Value

0.99 (0.98–1.01)

0.30 0.88

CONTEMPORARY TREND OF CIED INFECTION. Cor-

Male

1.04 (0.62–1.75)

responding to the increase in CIED implantation rate

Hypertension

0.66 (0.39–1.14)

0.14

in the recent 3 decades, the incidence of CIED infec-

Hyperlipemia

1.10 (0.64–1.89)

0.73

tion increased to 4.2 per 1,000 person-years, which

Congestive heart failure

1.54 (0.91–2.61)

0.11

was higher than that between 1975 and 2004, when a

Cardiomyopathy

1.17 (0.67–2.06)

0.58

previous cohort study reported 1.9 per 1000 personyears (12). The most significant rise in the device implantation rate occurred in the late 1990s and early

Atrial fibrillation

1.38 (0.83–2.32)

0.22

Diabetes mellitus

2.12 (0.99–4.56)

0.05

Coronary artery disease

1.29 (0.76–2.20)

0.35

HR (95% CI)

p Value

2.42 (1.12–5.23)

0.03

2.16 (1.04–4.49)

0.04

Myocardial infarction

1.09 (0.63–1.89)

0.76

Percutaneous transluminal coronary angioplasty

1.00 (0.49–2.04)

0.99

Peri-vascular disease

1.26 (0.73–2.19)

0.40

Cerebrovascular disease

0.99 (0.56–1.75)

0.98

Stroke

1.03 (0.56–1.92)

0.92

Transient ischemic attack

1.08 (0.53–2.20)

0.84

of the REP database, the incidence of CIED infection

Valvular heart disease

2.59 (1.34–5.00)

0.005

in patients with ICDs was significantly higher than

Chronic kidney disease

1.29 (0.70–2.39)

0.42

those with PPMs (12). Although another large study

Chronic obstructive pulmonary disease

1.12 (0.62–2.02)

0.71

Malignancy

0.99 (0.57–1.70)

0.96

Obesity (BMI >25 kg/m2)

1.02 (0.61–1.71)

0.93

Sepsis

1.06 (0.38–2.95)

0.91

and CRT recipients compared with PPM recipients.

Endocarditis

2.40 (0.96–6.01)

0.06

1.49 (0.54–4.13)

0.44

The increase in the proportion of ICD and CRT im-

ICD*

2.96 (1.71–5.12)

<0.001

3.09 (1.78–5.35)

<0.001

plantations associated with a larger device, more

CRT*

3.76 (1.56–9.07)

<0.01

3.42 (1.41–8.32)

<0.01

2000s, and was likely associated with a significant increase in PPM implantation with population aging and ICD implantation for sudden death prevention; this was consistent with the findings of a national survey in 2006 (3). Similar to the previous publication

found no difference in the cardiac device infection rate between the ICD and PPM groups (18), our analysis revealed a 3-fold increase in infection rate in ICD

transvenous leads, and longer procedure time may have contributed to the increase in the incidence of infection.

*Permanent pacemakers as reference. CI ¼ confidence interval; HR ¼ hazard ratio; other abbreviations as in Table 1.

1078

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

comorbidities, set the stage for higher rates of CIED

ICD and/or CRT, especially for the third procedure.

infection (16,19). Although 2 previous studies identi-

Enhanced prophylactic measures such as using

fied younger age as one of the risk factors associated

an

with the PPM infection (20,21), our data did not reveal

when

any significant difference in the infection incidence

procedures.

antibacterial these

envelope

patients

would

undergo

be

beneficial

repeated

CIED

between the different age groups before 2008. However, the infection incidence in the 50- to 70-year old

PREDICTORS OF CIED INFECTION AND SURVIVAL. Some

group became significantly higher than the other 2

earlier research suggested that the CIED infection

age groups in 2009 to 2015. This age group received

could lead to a higher mortality compared with those

ICDs and CRT more often and were more likely

without infection (13,24,26). However, we did not

exposed to subsequent generator replacements than

observe an association between CIED infection and a

the older age group. This might explain a higher

higher long-term mortality rate, which could be

incidence of infection in the era of advanced

explained by a small incidence of a <3% infection rate

device therapy. Caution should be undertaken when

and a long duration of follow-up (27). Risk factors for

this

CIED infection were described in several earlier pub-

group

of

patients

undergoes

subsequent

lications (22,28–32). We did not observe an associa-

procedures. INFECTION AND REPEATED CIED PROCEDURES. This

study showed that most CIED infection events occurred after the de novo implantation (>50%). It reinforced the idea of taking universal precautions, such as using proper sterile techniques and prophylactic intravenous antibiotics during device implantation. Previous studies suggested that repeated procedures were associated with an increased risk of infection

(5,13,22,23).

The

cumulative

infection

incidence of CIED replacements at 1-year postimplantation was twice as high as that of initial implantations (2.4% vs. 1.2%) (13). We found the incidence of infection following the second procedure was similar to that after the initial implantation, but substantially increased 8-fold after the third procedure at 15 years of follow-up. The difference between our findings and previous reports might be due to our unique longer follow-up duration. The

tion of age, sex, chronic kidney disease, chronic obstructive pulmonary disease, malignancy, obesity, or heart failure with an increased risk of CIED infection. However, presence of diabetes mellitus and valvular heart disease were associated with a 2-fold increase in risk of infection. Diabetes mellitus is considered to be in immunocompromised state that predisposes patients to infection with or without associated

invasive

procedures

(33–35).

Native

valvular disease and prosthetic heart valves are known risk factors for infective endocarditis, regardless of the causes of valvular diseases or the type of valve (34,36). Our analysis, during 30 years of followup,

provided

supportive

data

to

these

earlier

observations. The previously reported risk factors associated with CIED infection, such as heart failure and chronic kidney disease, were not significant in our study, which was likely due to a small number of infections overall.

incidence of infection in the PPM group remained similar regardless of the number of procedures,

STUDY LIMITATIONS. Although our study included a

whereas the ICD group had a greater risk of infec-

large population-based cohort with long-term follow-

tion associated with multiple CIED procedures. The

up, it had several limitations. First, as in any obser-

most commonly repeated procedures were generator

vational study, we could not rule out the effect of

replacement and device upgrade (24). Generator

residual confounding due to unmeasured variables.

change and the device upgrade procedure might

Second, by analyzing the presence of an association

interrupt the equilibrium between the organisms,

with a specific procedure type after initial implanta-

which could colonize a pocket after initial implan-

tion and infection, the confounding of other proced-

tation without clinical manifestations, and the host

ure types might exist. Third, our study cohort

immune response (24,25). It was conceivable that re-

primarily included a single community of mostly

opening a relatively larger device pocket capsule or

white ethnicity, and patients were treated at only 2

prolonged pocket exposure time for adding a new

hospitals with a small number of infection events,

lead might predispose patients to CIED infection.

which might not be reflective of other centers and

This finding suggested that it was important to

patient populations. Finally, there was a possibility of

consider risks and/or benefits in upgrades or need

underestimating the incidence of CIED infection by

for generator replacement in patients who received

using ICD-9-CM codes.

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

CONCLUSIONS

PERSPECTIVES

This population-based study described a trend of

COMPETENCY IN MEDICAL KNOWLEDGE: This study

increasing incidence of CIED infection over the last 2

evaluated CIED infection trends in a population-based cohort

decades. In keeping with improved survival, lead

with an impressive longitudinal follow-up. Although the overall

dwell time, and repeated procedures for generator

incidence of CIED infection was low for new implantation, the

changes and device upgrades in patients who received

risk of infection was significantly increased for generator

transvenous CIEDs, enhanced prophylactic measures

replacement or device upgrade, especially the third procedure on

to reduce the incidence of CIED infection are imper-

the device pocket.

ative for successive ICD and CRT procedures.

TRANSLATIONAL OUTLOOK: These results suggest the

ADDRESS FOR CORRESPONDENCE: Dr. Yong-Mei

importance of using more effective prophylactic strategies to

Cha, Department of Cardiovascular Diseases, Mayo

reduce CIED infection in high-risk patients.

Clinic, 200 First Street SW, Rochester, Minnesota 55905. E-mail: [email protected].

REFERENCES 1. Greenspon AJ, Patel JD, Lau E, et al. 16-year

organization),

Society

19. Uslan DZ, Gleva MJ, Warren DK, et al. Car-

trends in the infection burden for pacemakers and implantable cardioverter-defibrillators in the United States 1993 to 2008. J Am Coll Cardiol 2011;58:1001–6.

(BHRS), British Cardiovascular Society (BCS), British Heart Valve Society (BHVS) and British Society for Echocardiography (BSE). J Antimicrob Chemother 2015;70:325–59.

diovascular implantable electronic device replacement infections and prevention: results from the REPLACE Registry. Pacing Clin Electrophysiol 2012;35:81–7.

2. Voigt A, Shalaby A, Saba S. Continued rise in

10. Boggan JC, Baker AW, Lewis SS, et al. An automated surveillance strategy to identify infectious complications after cardiac implantable electronic device procedures. Open Forum Infect

20. Johansen JB, Jorgensen OD, Moller M,

rates of cardiovascular implantable electronic device infections in the United States: temporal trends and causative insights. Pacing Clin Electrophysiol 2010;33:414–9.

British

Heart

Rhythm

Dise 2015;2:ofv128.

3. Voigt A, Shalaby A, Saba S. Rising rates of cardiac rhythm management device infections in the

11. Melton LJ, 3rd. History of the Rochester Epide-

United States: 1996 through 2003. J Am Coll Cardiol 2006;48:590–1.

12. Uslan DZ, Sohail MR, St Sauver JL, et al. Per-

4. Cabell CH, Heidenreich PA, Chu VH, et al. Increasing rates of cardiac device infections among Medicare beneficiaries: 1990-1999. Am Heart J 2004;147:582–6. 5. Nery PB, Fernandes R, Nair GM, et al. Devicerelated infection among patients with pacemakers and implantable defibrillators: incidence, risk factors, and consequences. J Cardiovasc Electro-

miology Project. Mayo Clin Proc 1996;71:266–74.

manent pacemaker and implantable cardioverterdefibrillator infection: a population-based study. Arch Int Med 2007;167:669–75. 13. Sohail MR, Eby EL, Ryan MP, Gunnarsson C, Wright LA, Greenspon AJ. Incidence, treatment intensity, and incremental annual expenditures for patients experiencing a cardiac implantable electronic device infection: evidence from a large US

physiol 2010;21:786–90.

payer database 1-year post implantation. Circ Arrhythm Electrophysiol 2016;9.

6. Landolina M, Gasparini M, Lunati M, et al. Longterm complications related to biventricular defibrillator implantation: rate of surgical revisions and impact on survival: insights from the Italian

14. Chamis AL, Peterson GE, Cabell CH, et al. Staphylococcus aureus bacteremia in patients with permanent pacemakers or implantable cardioverterdefibrillators. Circulation 2001;104:1029–33.

Clinical Service Database. Circulation 2011;123: 2526–35.

15. Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 2000;30:

7. Maytin M, Wilkoff BL, Brunner M, et al. Multicenter experience with extraction of the Riata/ Riata ST ICD lead. Heart Rhythm 2014;11:1613–8. 8. Gold MR, Peters RW, Johnson JW, Shorofsky SR. Complications associated with pectoral cardioverter-defibrillator implantation: comparison of subcutaneous and submuscular approaches. Worldwide Jewel Investigators. J Am Coll Cardiol 1996;28:1278–82.

633–8.

Arnsbo P, Mortensen PT, Nielsen JC. Infection after pacemaker implantation: infection rates and risk factors associated with infection in a population-based cohort study of 46299 consecutive patients. Eur Heart J 2011;32:991–8. 21. Margey R, McCann H, Blake G, et al. Contemporary management of and outcomes from cardiac device related infections. Europace 2010;12: 64–70. 22. Klug D, Balde M, Pavin D, et al. Risk factors related to infections of implanted pacemakers and cardioverter-defibrillators: results of a large prospective study. Circulation 2007;116:1349–55. 23. Catanchin A, Murdock CJ, Athan E. Pacemaker infections: a 10-year experience. Heart Lung Circ 2007;16:434–9. 24. Prutkin JM, Reynolds MR, Bao H, et al. Rates of and factors associated with infection in 200 909 Medicare implantable cardioverterdefibrillator implants: results from the National Cardiovascular Data Registry. Circulation 2014; 130:1037–43. 25. Blaser MJ, Kirschner D. The equilibria that allow bacterial persistence in human hosts. Nature 2007;449:843–9.

16. Kusumoto FM, Schoenfeld MH, Wilkoff BL, et al. 2017 HRS expert consensus statement on cardiovascular implantable electronic device lead management and extraction. Heart Rhythm 2017; 14:e503–51. 17. DeSimone DC, Sohail MR. Approach to diagnosis of cardiovascular implantable-electronicdevice infection. J Clin Microbiol 2018;56: e01683–17.

9. Sandoe JA, Barlow G, Chambers JB, et al. Guidelines for the diagnosis, prevention and management of implantable cardiac electronic device infection. Report of a joint Working

18. Lekkerkerker JC, van Nieuwkoop C, Trines SA, et al. Risk factors and time delay associated with

Party project on behalf of the British Society for Antimicrobial Chemotherapy (BSAC, host

cardiac device infections: Leiden device registry. Heart 2009;95:715–20.

26. Sohail MR, Henrikson CA, Braid-Forbes MJ, Forbes KF, Lerner DJ. Mortality and cost associated with cardiovascular implantable electronic device infections. Arch Intern Med 2011;171: 1821–8. 27. Deharo JC, Quatre A, Mancini J, et al. Longterm outcomes following infection of cardiac implantable electronic devices: a prospective matched cohort study. Heart 2012;98:724–31. 28. Sohail MR, Uslan DZ, Khan AH, et al. Risk factor analysis of permanent pacemaker infection. Clin Infect Dis 2007;45:166–73.

1079

1080

Dai et al.

JACC: CLINICAL ELECTROPHYSIOLOGY VOL. 5, NO. 9, 2019 SEPTEMBER 2019:1071–80

Trends in CIED Infection

29. Le KY, Sohail MR, Friedman PA, et al. Clinical predictors of cardiovascular implantable electronic device-related infective endocarditis. Pacing Clin Electrophysiol 2011;34:450–9. 30. Sohail MR, Hussain S, Le KY, et al. Risk factors associated with early- versus late onset implantable cardioverter-defibrillator infections. J Interv Card Electrophysiol 2011;31:171–83. 31. Romeyer-Bouchard C, Da Costa A, Dauphinot V, et al. Prevalence and risk factors related to infections of cardiac resynchronization therapy devices. Eur Heart J 2010;31:203–10. 32. de Oliveira JC, Martinelli M, Nishioka SA, et al. Efficacy of antibiotic prophylaxis before the implantation of pacemakers and cardioverter-

defibrillators: results of a large, prospective, randomized, double-blinded, placebo-controlled trial. Circ Arrhythm Electrophysiol 2009;2:29–34. 33. McMahon MM, Bistrian BR. Host defenses and susceptibility to infection in patients with diabetes mellitus. Infect Dis Clin N Am 1995;9:1–9.

university medical center. Clin Infect Dis 1998;26: 1111–6. 36. Iung B, Vahanian A. Epidemiology of acquired valvular heart disease. Can J Cardiol 2014;30: 962–70.

34. Herce B, Nazeyrollas P, Lesaffre F, et al. Risk factors for infection of implantable cardiac devices: data from a registry of 2496 patients. Europace 2013;15:66–70. 35. Spinler SA, Nawarskas JJ, Foote EF, Sabapathi D, Connors JE, Marchlinski FE. Clinical presentation and analysis of risk factors for infectious complications of implantable cardioverter-defibrillator implantations at a

KEY WORDS cardiovascular implantable electronic devices infection, populationbased study, trends

A PPE NDI X For supplemental tables and a figure, please see the online version of this paper.