Predictors of Long-Term Infections After Cardiac Implantable Electronic Device Surgery　― Utility of Novel PADIT and PACE DRAP Scores ―

Sylwia Sławek-Szmyt; Aleksander Araszkiewicz; Marek Grygier; Krzysztof Szmyt; Lidia Chmielewska-Michalak; Wojciech Seniuk; Michał Waśniewski; Tomasz Smukowski; Maciej Lesiak; Przemysław Mitkowski

doi:10.1253/circj.CJ-20-0305

Abstract

Background: Cardiac implantable electronic device-related infections (CDI) are of increasing importance and involve substantial healthcare resources. This study aimed to evaluate potential CDI risk factors and the utility of the novel PADIT and PACE DRAP scores to predict CDI.

Methods and Results: The study group included 1,000 consecutive patients undergoing implantable cardioverter-defibrillator (ICD) or cardiac resynchronization therapy (CRT) surgery. Patients’ and procedural characteristics were collected. CDI occurrence was assessed during 1-year follow-up. Moreover, if periprocedural significant pocket hematoma (SPH) occurred, the maximal volume was calculated based on ultrasonographic measurements and ABC/2 formula. The overall incidence of CDI was 1.8%. In the multivariable regression analysis independent CDI risk factors were: age >75 years (odds ratio [OR]: 5.93; 95% confidence interval [CI]: 1.77–19.84), system upgrade procedure (OR: 6.46; CI: 1.94–21.44), procedure duration >1 h (OR: 13.96; CI: 4.40–44.25), presence of SPH (OR: 4.95; CI: 1.62–15.13) and reintervention within 1 month (OR: 16.29; CI: 3.14–84.50). The PACE DRAP score had higher discrimination of CDI incidence (area under curve [AUC] 0.72) as compared with the PADIT score (AUC 0.63).

Conclusions: We identified 5 independent risk factors of CDI development. Our study also showed that the PACE DRAP score was better able to identify patients at high risk of CDI than the PADIT score.

The number of cardiac implantable electronic device (CIED) procedures in patients with different cardiac rhythm disorders or heart failure has significantly increased worldwide over the past decades and has continued to expand with the growing indications for CIED and population aging. Despite significant progress in technology development and growing operator experience, the rate of device-related complications still remains high.

CIED-related infections (CDI) are the most serious complication associated with increased mortality and involve substantial healthcare resources and costs.¹^,² The reported CDI occurrence varies between 0.68% and 2.2% of CIED procedures, with significantly higher CDI incidence for implantable cardioverter-defibrillator (ICD) or cardiac resynchronization therapy (CRT) procedures (up to 4.8%).²^–⁵ Previous studies report several patient-related and procedure-related risk factors for CDI development, but many of them were retrospective analyses with inconsistent CDI definitions.⁴^,⁶^,⁷ Moreover, only a few of the previous reports focus on ICD or CRT recipients particularly.⁸ It seems reasonable that a larger size generator may predispose to skin necrosis and subsequent CDI development.

Recently, a novel infection risk score, the PADIT score, was proposed for CIED recipients. The PADIT score consists of 5 independent factors related to patients’ clinical state and procedural details (Table 1).⁹

Table 1. The PACE DRAP and PADIT Scores Characteristics

PACE DRAP SCORE				PADIT SCORE
Letter designation	Risk factor	Definition	Points	Letter designation	Risk factor	Definition	Points
P	Prosthesis	Biological/mechanical valvular prosthesis	+2	P	Prior procedures	No. of previous procedures 1 ≥2	+1 +4
A	Arterial hypertension uncontrolled	Blood pressure ≥160/100 mmHg	+2	A	Age	<60 years 60–69 years	+2 +1
C	Cancer	Any malignancy diagnosed or treated within past 5 years	+2	D	Depressed eGFR <30 mL/min/1.73 m²	Renal insufficiency	+1
E	Elderly	Age ≥75 years	+2	I	Immunocompromised	Receiving therapy that suppresses resistance to infection (e.g., immunosuppression, high- dose steroids) or having a disease that suppress resistance to infection (e.g., leukemia, HIV infection)	+3
D	Device type	CRT/ICD surgery	+2	T	Procedure type	ICD CRT Revision/upgrade	+2 +4 +5
R	Renal failure	eGFR <60 mL/min/1.73 m²	+1
A	Antiplatelets	Clopidogrel Ticagrelor	+2 +3
P	Procedure type	System upgrade	+2

CRT, cardiac resynchronization therapy; eGFR, estimated glomerular filtration rate; ICD, implantable cardioverter-defibrillator.

On the other hand, bleeding complications, including significant pocket hematoma (SPH), related to CIED surgery may have very significant sequelae for patients. Many risk factors and controversial reports for SPH and CDI associations are declared in the literature,¹⁰^,¹¹ but although a previous multicenter randomized study indicated that SPH occurrence increases the risk of subsequent infection 7-fold, the definition of SPH was not standardized, which might cause some bias.¹⁰ Importantly, during the past 5 years the periprocedural antithrombotic strategy has changed significantly, particularly by avoiding bridging therapy with low molecular weight heparins.¹² Nonetheless, the great majority of past studies reporting on long-term CDI frequency and predictors have been in the previous clinical setting.¹⁰

What is interesting is the first bleeding score with the PACE DRAP acronym newly developed for CIED recipients to predict SPH incidence. The cutoff value for high risk of SPH is established as ≥6 points (Table 1).¹³

The aim of the present study was to identify potential risk factors of subsequent CDI development. We aimed also to evaluate the association between SPH and its estimated volume on the risk of CDI occurrence. Another purpose of the study was to compare the PADIT and PACE DRAP risk scores in their capacity to predict CDI in a real-world cohort of unselected CIED recipients.

Methods

This prospective cohort study was conducted between January 2016 and January 2020 in a tertiary-care cardiology center. A total 1,015 consecutive patients undergoing ICD or CRT (both CRT-pacemakers [CRT-P] and CRT-defibrillators [CRT-D]) implantation, system replacement or upgrade procedure (replacement with a device with an additional lead to upgrade the function of the previous system) were enrolled into the study. We included only ICD or CRT devices because of their similar generator dimensions (significantly larger than pacemakers). Exclusion criteria were pacemaker surgery, pregnancy, inability to sign the informed consent, participation in another study, lead and/or device extraction procedure, and geographic inaccessibility for follow-up. However, 15 patients were lost during follow-up and were excluded, making the final study population of 1,000 ICD/CRT recipients. The study protocol was in accordance with the Declaration of Helsinki and was approved by the local ethics committee (approval no. 613/15).

All patients provided written informed consent prior to enrollment in the study. The privacy of the patients was protected by anonymization of all data.

Patients

We collected demographic and medical data, including comorbidities, indication for ICD/CRT surgery, usage and type of chronic antiplatelet or antithrombotic therapy, and routine laboratory test findings, as well as procedural details and complication management during primary hospitalization. During the 1-year follow-up, collected data included hospitalization information for CDI, evidence for the infection, culture and microorganism details, management of the infection, and complications from the infection or its management.

Anticoagulation therapy (ACT) was managed according to current European Heart Rhythm Association (EHRA) recommendations.¹² In patients with intermediate or high risk of thromboembolic events, vitamin K antagonist (VKA) treatment was continued with international normalized ratio (INR) ≤3.0 (≤3.5 in patients with mechanical valve prosthesis), but in patients with low thromboembolic risk, VKA was interrupted periprocedurally. Direct oral anticoagulant (DOAC) treatment was interrupted perioperatively for a period depending on the drug type and renal function for 24–36 h. Antiplatelet therapy (APT) was uninterrupted.¹²

CIED Procedures

All device procedures were performed in a standard way with prepectoral subcutaneous pocket formation and transvenous lead placement by subclavian or axillary vein access. All right atrial and right ventricular leads were positioned in the right auricular appendage and right ventricular apex, respectively. The left ventricular leads were positioned in the lateral, posterolateral or anterior cardiac vein at the operator’s discretion. In the case of device replacement, the existing capsule was removed with the use of electrocautery at the time of procedure. All procedures were performed by 3 experienced electrophysiology specialists with a high annual number of procedures.

Furthermore, in preparation for the procedure, all patients received a prophylactic dose of antibiotic intravenously (single dose of cefazoline or single dose of clindamycin in the case of allergy to penicillin). No antibiotics were administered after the primary procedure. In the current study population, no prohemostatic agents, pressure dressings or sandbags were used to prevent pocket complications.

Endpoints and Definitions

The primary outcomes of the study were the occurrence of SPH within 1 month after ICD/CRT procedure and the incidence of CDI during 12 month follow-up.

SPH Definition

Pocket bleeding complications were classified according to previous recommendations.¹⁴

SPH was defined as a swelling and painful mass extending to the margins of the generator, prolonging hospitalization for at least 24 h after the CIED procedure or needing interruption of ACT/APT or requiring evacuation. SPH was treated conservatively with prolonged manual compression followed by pressure bandage dressing applied for 24 h.

If the SPH required evacuation, the procedure was performed under aseptic conditions in the operating room. The occurrence of any pocket bleeding complications was assessed 24 h after the CIED procedure and every 7 days during the first month after CIED surgery on outpatient visit.

If SPH was diagnosed, ultrasound examination (ClearVue 850 Ultrasound Machine, Philips, Amsterdam, The Netherlands) was performed immediately, and then repeatedly every 3–4 days up to resolution or evacuation. The typical ultrasonographic SPH appearance included a circumscribed hypoechoic or anechoic mass, partially compressible without calcification. Hematomas were characterized by an ellipsoid or lentiform shape. In some cases, an additional irregular pattern within the surrounding tissues was seen as a result of interstitial bleeding. To precisely estimate SPH volume 3 basic dimensions were measured ultrasonographically: maximal horizontal diameter (A), maximal vertical diameter (B) and maximal longitudinal diameter (C). All ultrasound measurements at the same point of time were performed twice by the same investigator, and the mean value of the maximal hematoma dimensions was used in further investigations. Details are presented in Figure 1. Next, the hematoma volume was calculated according to a simplified formula for the volume of an ellipsoid: ABC/2 formula. This ABC method was adapted from a widely used intracranial hematoma volume estimation formula.¹⁵^,¹⁶ In the case of SPH occurrence prophylactic antibiotic therapy with amoxicillin/clavulanic acid twice daily for 7 days was administered.

Figure 1.

Calculation of the volume of significant pocket hematoma (SPH). (A) Macroscopic image of SPH. (B) Scheme of SPH volume calculation with use of the formula for volume of an ellipsoid. SPH dimensions: A – maximal horizontal diameter (cm); B – maximal vertical diameter (cm); C – maximal longitudinal diameter (cm). (C–E) Ultrasonographic measurement of each SPH dimension: (C) Maximal horizontal diameter, (D) maximal vertical diameter and (E) maximal longitudinal diameter.

CDI

CDI was defined as the presence of local signs of inflammation at the device pocket, including erythema, warmth, tenderness, fluctuance, wound dehiscence, erosion or purulent drainage. CIED-related endocarditis (CDRIE) was diagnosed when the modified Duke’s criteria for infective endocarditis were met by the presence of valvular or lead vegetations confirmed by echocardiographic assessment.¹⁷ All CDI-affected patients were treated with appropriate intravenous antibiotic therapy and complete system removal whenever feasible.

CDI Risk Assessment The risk of CDI in each patient was assessed using 2 tools: PADIT score (prior procedures, age, depressed glomerular filtration rate, immunocompromised, type of procedure) and PACE DRAP score (prosthesis, age, cancer, elderly, device type, renal failure, antiplatelet drugs, procedure type).⁹^,¹³ Details are presented in Table 1.

Statistical Analysis

Patients’ characteristics are expressed as frequency/percentage for categorical variables and median/interquartile range (IQR) for continuous variables (none of the assessed parameters had normal distribution according to the Shapiro-Wilk test). Categorical variables were compared using the two-tailed Fisher’s exact test or χ²-test as appropriate, and continuous variables were analyzed using the Mann-Whitney U test. A two-tailed α of 0.05 and a P-value of <0.05 were considered statistically significant. Logistic regression analyses were used to identify the independent predictors of CDI. The multicollinearity of potential predictors was tested with a variance inflation factors (VIF) method and predictors with a VIF value of at least 4 were excluded from the final multivariable regression model. Receiver operating characteristic curve (ROC) analysis was performed to determine the predictive value of the PACE DRAP and PADIT scores and for evaluation of the regression models. The statistical analyses were performed using Statistica 13.7 version (StatSoft, Inc., Tulsa, OK, USA).

Results

CDI Incidence

Within the 1 year of follow-up a total 18 (1.8%) cases of CDI were identified. Detailed baseline characteristics of patients with and without CDI are presented in Table 2. Patients with CDI were significantly older (median age 76 years, IQR: 72–79) as compared with patients without CDI (median age 69 years, IQR: 60–76). Patients with CDI significantly more often had a higher functional class (New York Heart Association [NYHA] III–IV), diabetes mellitus, renal failure (estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m²) with increased serum creatinine concentration, but also nontherapeutic INR (INR <2.0) and lower hemoglobin concentration on the day of the procedure. However, coronary artery disease, as well as DOACs usage and single-APT (SAPT), was less frequent in the CDI group in comparison with the no-CDI group. The median duration of the procedure was significantly longer in patients with CDI (60 min, IQR: 43–65 min) in comparison with the no-CDI patients (34 min, IQR: 21–45 min). Upgrade procedure was also more frequent in the CDI group (38.9%) than in the no-CDI group (11.7%). The frequencies of SPH, pneumothorax and reintervention within the first month after CIED procedure were also significantly higher in the CDI group. Among all CDI, the device-related local infection rate was 83.3%, and that of CDRIE was 16.7%. Details are given in Table 3.

Table 2. Baseline Patient and Procedural Characteristics

	CDI group (n=18)	No CDI group (n=982)	P value
Age median (IQR) (years)	76 (72–79)	69 (60–76)	0.0037
Female/male	7 (38.9)/11 (61.1)	371 (37.8)/611 (62.2)	0.9
BMI median (IQR) (kg/m²)	26.1 (22.9–28.0)	26.8 (24.2–30.5)	0.057
Procedure duration median (IQR) (min)	60 (43–65)	34 (21–45)	<0.0001
Coronary artery disease	11 (61.1)	816 (83.1)	0.024
NYHA functional class			0.049
I–II	7 (38.9)	615 (62.6)
III–IV	11 (61.1)	367 (37.4)
Arterial hypertension	12 (66.7)	480 (48.9)	0.16
Atrial fibrillation	10 (55.6)	362 (37.3)	0.14
Diabetes mellitus	8 (44.4)	230 (23.4)	0.049
Renal failure (eGFR <60 mL/min/1.73 m²)	10 (55.6)	305 (31.1)	0.038
Previous stroke
Ischemic/hemorrhagic	2 (11.1)/–	63 (6.4)/6 (0.61)	0.4
Prosthetic valve	3 (16.7)	50 (5.1)	0.65
Smoking			0.2
Active	4 (23.5)	157 (16.3)
Past	1 (5.9)	217 (22.5)
Never	13 (72.2)	608 (61.9)
Immunocompromised	3 (16.7)	15 (1.5)	0.09
Malignancy	1 (5.6)	11 (1.1)
Breast cancer	–	2 (0.2)	0.2
Prostate cancer	1 (5.6)	4 (0.4)	0.2
Colorectal cancer	–	4 (0.4)
Myeloproliferative leukemia	–	1 (0.1)
Oral anticoagulation	12 (66.7)	408 (41.5)	0.051
VKA INR ≥2.0 on procedure day	2 (11.1)	103 (10.5)	0.75
VKA INR <2.0 on procedure day	5 (27.8)	95 (9.7)	0.027
DOACs	2 (11.1)	132 (13.4)	0.001
Antiplatelets	6 (33.3)	440 (44.8)	0.65
SAPT	–	208 (21.2)	0.034
DAPT	3 (16.7)	154 (16.1)	0.05
TAT	2 (11.1)	60 (6.1)	0.3
DAT	1 (5.6)	18 (1.8)	0.3
Type of device
ICD	14 (77.8)	643 (65.5)	0.27
CRT	4 (22.2)	339 (34.5)	0.33
Type of procedure
De novo implantation	10 (55.6)	519 (58.2)	0.81
Replacement	1 (5.6)	348 (35.4)	0.006
System upgrade	7 (38.9)	115 (11.7)	0.00047
LVEF, Median (IQR) (%)	35 (25–35)	30 (25–35)	0.1136
Serum creatinine concentration on procedure day, Median (IQR) (μmol/L)	145.5 (93.7–198.0)	95.0 (77.6–116.0)	0.0022
GFR on procedure day, Median (IQR) (mL/min)	54.5 (33–80.0)	55.0 (70.0–89.0)	0.04
Hemoglobin concentration on procedure day, Median (IQR) (mmol/L)	7.5 (6.2–8.5)	7.9 (8.6–9.2)	0.0003
White blood cells on procedure day, Median (IQR) (1×10³/μL)	7.92 (6.3–8.7)	6.42 (7.8–9.0)	0.7245
CRP level on procedure day, Median (IQR) (mg/L)	6.0 (4.8–7.7)	5.0 (3.1–7.3)	0.1331
INR level on procedure day, Median (IQR)	1.15 (1.0–2.1)	1.0 (1.0–1.2)	0.309
PADIT score, Median (IQR)	4.5 (4–8)	4 (3–6)	0.14
PACE DRAP score, Median (IQR)	8 (4–10)	4 (3–6)	0.0001
CHA₂DS₂-VASc score, Median (IQR)	5 (3–6)	3 (2–4)	0.067
SPH	11 (61.1)	36 (3.7)	<0.0001
SPH volume, Median (IQR) (cm³)	193.6 (327.6–367.2)	114.8 (96.7–164.7)	0.0005
Pneumothorax	2 (11.1)	17 (1.73)	0.044
Reintervention within 1 month	3 (16.7)	13 (1.32)	<0.0001

Unless indicated otherwise, data are given as n (%). BMI, body mass index; CDI, cardiac implantable electronic device-related infection; CHA₂DS₂-VASc score, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65–74 years, sex category; CRP, C-reactive protein; CRT, cardiac resynchronization therapy; DAPT, dual-antiplatelet therapy; DAT, double antithrombotic therapy; DOAC, direct oral anticoagulant; eGFR, estimated glomerular filtration rate; ICD, implantable cardioverter-defibrillator; INR, international normalized ratio; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; SAPT, single-antiplatelet therapy; SPH, significant pocket hematoma; TAT, triple antithrombotic therapy; VKA, vitamin K antagonist.

Table 3. Detailed Description of Patients With Cardiac Implantable Electronic Device-Related Infection

Patient ID	Sex	CDI Endocarditis/ pocket	Pathogen isolated	Previous SPH	Previous SPH volume (cm³)	PACE DRAP score	PADIT score	Outcome
1	M	Pocket	MSSA	Yes	327.6	9	12	Alive
2	M	Pocket	MRSA	Yes	336	4	2	Alive
3	M	Pocket	MRSE	No	–	10	9	Alive
4	F	Endocarditis	Enetrecoccus faecalis	Yes - evacuated	541.5	9	8	Not alive
5	F	Pocket	MRSE	No	–	2	7	Alive
6	F	Pocket	MSSA	No	–	10	4	Alive
7	F	Pocket	MSSA	No	–	2	4	Alive
8	F	Pocket	MRSA	Yes	99	6	3	Alive
9	M	Pocket	MRSE	Yes	225	3	2	Alive
10	M	Pocket	MRSE	Yes	378.7	11	8	Alive
11	F	Pocket	MRSA	No	–	6	7	Alive
12	M	Pocket	MRSE	No	–	4	4	Alive
13	M	Endocarditis	Staphylococcus haemolyticus	No	–	7	11	Alive
14	M	Pocket	MSSA	Yes	292.8	7	3	Alive
15	F	Pocket	MSSE	Yes	193.6	9	5	Alive
16	M	Pocket	negative	Yes	344.4	12	9	Alive
17	M	Endocarditis	MRSA	Yes	97	12	4	Alive
18	M	Pocket	negative	Yes	367.2	10	4	Alive

CDI, cardiac implantable electronic device-related infection; MRSA, methicillin-resistant Staphylococcus aureus; MRSE, methicillin-resistant Staphylococcus epidermidis; MSSA, methicillin-sensitive Staphylococcus aureus; MSSE, methicillin-sensitive Staphylococcus epidermidis; SPH, significant pocket hematoma.

SPH Volume

The overall frequency of SPH was 4.7% (47/1,000 study population). It was significantly more frequent in CDI patients (61.1%) than in patients without CDI (3.7%) (P<0.0001). Among the whole study population, 1 patient in the CDI group (5.6%) and 2 in the no-CDI group (0.2%) underwent SPH hematoma evacuation.

The SPH volume significantly differed between groups. The median SPH volume in CDI patients was 193.6 cm³ (IQR: 327.6–367.2 cm³), while in patients without infection the median SPH volume was 114.8 cm³ (IQR: 96.7–164.7 cm³). The great majority of SPHs (9/11) in CDI patients had an estimated volume >200 cm³, whereas in the no-CDI patients the estimated volume of SPHs was most frequently <200 cm³ (29/36). Detailed data regarding SPH volume are shown in Table 4.⁹^,¹⁸

Table 4. Detailed Frequency of Significant Bleeding Complications After Device Surgery

SPH volume (cm³)	All patients n=1,000 (100%)	CDI group n=18 (100%)	No CDI group n=982 (100%)	P value
No SPH	953 (95.3)	7 (38.9)	946 (96.3)	0.036
<100	17 (1.7)	2 (11.1)	15 (1.5)	0.035
100–200	15 (1.5)	1 (5.6)	14 (1.4)	0.24
200–300	9 (0.9)	2 (11.1)	7 (0.7)	0.01
>300	6 (0.6)	6 (33.3)	0 (0)	<0.000001

SPH, significant pocket hematoma.

PADIT and PACE DRAP Scores for CDI Prediction

Patients were classified according to their PADIT and PACE DRAP scores, and at the time of surgery the median PADIT score in the CDI group was 4.5 (IQR: 4–8) and the median PACE DRAP score was 8 (IQR: 4–10).

In this study population, the PACE DRAP score better discriminated between patients with high and low risk of CDI development (AUC 0.72), in comparison with the PADIT score (AUC 0.63). Details are presented in Table 5 and Figure 2.

Table 5. Cardiac Implantable Electronic Device-Related Infection Rate and Negative and Positive Predictive Values for Infection Development Based on the PADIT and PACE DRAP Risk Assessment Models

Risk score	n	CDI (n)	PPV (%)	NPV (%)	Sensitivity (%)	Specificity (%)	AUC
PADIT							0.63
High risk (≥7 points)	233	9	3.7	98.8	50.0	76.3
Non-high risk (<7 points)	749	9	1.8	–	100.0	0
PACE DRAP							0.72
High risk (≥6 points)	284	13	4.4	99.3	72.2	71.1
Non-high risk (<6 points)	698	5	1.8	–	27.8	28.9

AUC, area under curve; CDI, cardiac implantable electronic device-related infection; NPV, negative predictive value; PPV, positive predictive value.

Figure 2.

Receiver operating characteristic (ROC) curve analysis of the PACE DRAP and PADIT scores for prediction of CDI development within 1 year after device surgery. AUC, area under curve; CDI, cardiac implantable electronic device-related infection.

Risk Factors of CDI

The results of the univariate and multivariable analyses for all potential predictors are presented in Table 6. The following characteristics were identified as independent predictors of CDI in the multivariable regression model: age >75 years (OR: 5.93; CI: 1.77–19.84), system upgrade procedure (OR: 6.46; CI: 1.94–21.44), duration of surgery >1 h (OR: 13.96; CI: 4.40–44.25), presence of SPH (OR: 4.95; CI: 1.62–15.13) and early reintervention within 1 month after primary procedure (OR: 16.29; CI: 3.14–84.50).

Table 6. Univariate and Multivariable Analyses of Cardiac Implantable Electronic Device-Related Infections Risk Factors

Predictor	Univariate			Multivariable
Predictor	OR	95%CI	P value	OR	95%CI	P value
Age >75 years	8.56	2.79–26.23	0.0002	5.93	1.77–19.84	0.004
Female sex	1.05	0.40–2.73	0.92	–
Functional NYHA Class III–IV	2.63	1.01–6.85	0.047	–
Coronary artery disease	0.89	0.34–2.39	0.82	–
Renal failure (eGFR <60 mL/min/1.73 m²)	6.79	2.64–17.42	0.000007	–
Atrial fibrillation	2.10	0.82–5.38	0.12	–
Diabetes mellitus	2.26	0.96–5.21	0.06	–
Arterial hypertension	1.64	0.63–4.26	0.31	–
Prosthetic valve (biological/mechanical)	5.3	1.68–16.75	0.0043	–
Previous stroke	1.41	0.37–5.40	0.62	–
Immunocompromised	3.17	0.89–11.25	0.07	–
Malignancy	5.2	0.63–42.51	0.12	–
VKA INR >2.0	1.17	0.26–5.15	0.84	–
VKA INR <2.0	1.71	0.49–6.00	0.40	–
TAT	1.89	0.49–6.00	0.40	–
DAT	3.15	0.39–24.96	0.28	–
DOAC	1.84	0.78–4.35	0.16	–
DAPT	3.42	1.31–8.96	0.012	–
CRT device	0.54	0.18–166	0.28	–
System upgrade procedure	4.7	1.77–12.25	0.002	6.46	1.94–21.44	0.002
De novo implantation	0.89	0.35–2.28	0.81	–
Procedure duration >1 h	10.58	4.03–27.79	0.000002	13.96	4.40–44.25	<0.000001
SPH	41.30	15.12–112.74	<0.000001	4.95	1.62–15.13	0.005
SPH volume ≥200 cm³	53.73	16.96–170.26	<0.000001	–
Pneumothorax	11.34	3.00–42.84	0.0003	–
Early reintervention (within 1 month)	14.90	3.84–57.80	<0.000001	16.29	3.14–84.50	0.0009

CI, confidence interval; OR, odds ratiost. Other abbreviations as in Table 2.

Our multivariable final model reached a high value of AUC (0.95) with P-value <0.000001.

Discussion

CDI Incidence

In the current study, CDI occurred in 1.8% of ICD/CRT recipients, which is similar to recently published data, but lower than in other reports.³^,⁴^,⁷^,⁸^,¹⁰^,¹⁸^,¹⁹ It should be pointed out that available literature data regarding CDI rate vary considerably depending on the definition of CDI, device and procedure type as well the duration of follow-up.²⁰

Previous studies report that the risk of CDI is the highest for CRT procedures (1.1–4.8%) followed by ICD (0.7–3.3%) and pacemaker surgery (0.5–1.19%).⁴^,⁵^,⁷ The increased rate of CRT device infections is associated with advanced device complexity and time-consuming procedure, but also patient multimorbidity is of great importance.²⁰ In our study population, all ICD and CRT procedures were performed by experienced operators with a high annual number of procedures, which probably influenced the lower CDI frequency.

Nonetheless, the CDI incidence reported here is related to the length of follow-up (1 year). Previous studies have indicated that the risk of CDI is highest within the first year after CIED surgery.⁴^,⁶ However, Jędrzejczyk-Patej et al⁸ demonstrated that during a median 40 months’ follow-up infective complications occurred in 44.7% within the first year of observation, but 55.3% of CDI cases were detected during the late observation period.⁸

Predictive Ability of PADIT and PACE DRAP Scores

Recently, Birnie et al⁹ developed a new risk score with the PADIT acronym for prediction of hospitalization due to device infection. They identified patients with at least 7 points in the PADIT score as having a high risk of CDI (4.33%). We previously developed a novel PACE DRAP risk prediction model that identified patients at high risk of significant bleeding complications (especially SPH) after CIED surgery. Scoring at least 6 points in PACE DRAP indicates a high risk of SPH.¹³ However, we hypothesized that our previously developed and published elsewhere PACE DRAP risk model would be able to accurately categorize patients into high and low predicted CDI risk groups because of a prior proven association between SPH occurrence and CDI development.¹⁰ In the present study, we compared, for the first time, the novel PADIT and PACE DRAP scores for predicting bleeding events in a clinical cohort undergoing CIED surgery. Importantly, we demonstrated a clear advantage of the PACE DRAP score, with AUC 0.72, over the PADIT score, with AUC 0.63. The obtained CDI predictive ability for the PACE DRAP score was more satisfactory, which we attribute to several factors. Firstly, in the present study there was a higher prevalence of the factors included in PACE DRAP score, such as older age (median age of CDI group: 76 years with IQR 72–79 years vs. 69 (IQR 60–76) years in no-CDI group) and renal failure (55.6% in CDI group and 31.1%, in no-CDI group). Secondly, we included only patients who had undergone ICD or CRT surgery with an estimated survival of more than 1 year and without any pacemaker procedures. Therefore, there was a low percentage of patients with active malignancy, which is a factor given 3 points weight in the PADIT score compared with 2 points weight in PACE DRAP.

CDI Risk Factors

Previous studies have identified different host-related and procedure-related risk factors of CDI. Endstage renal disease, heart failure, diabetes mellitus, and anticoagulation treatment, as well as procedure duration, postprocedural SPH and system replacement/upgrade procedures are among the most important of the listed factors.²⁰

In the present study only, ICD or CRT recipients were included, enabling precise evaluation of predictors in this population. Older age (>75 years old) independently predicted increased risk of CDI, which is in contrast to previous studies that found younger age as a CDI risk factor.³^,⁹^,²⁰ However, advanced age might be associated with CDI because of the patients’ multiple comorbidities and frailty.¹¹^,²¹ Furthermore, prior reports are limited to data obtained from patients with CDI and system removal, but older patients treated conservatively were not often enrolled.³^,⁴

Our study indicated that procedure duration longer than 1 h is an independent CDI risk factor. It is suggested that longer operating time is related to bacterial contamination.²⁰^,²² Moreover, these finding indicate the great importance of operator experience and the necessity of maintaining a high annual number of operations.

Other significant CDI predictors were system upgrade and early reintervention procedures, which was in line with previous reports.⁴^,²⁰ One of the potential mechanisms might be bacterial contamination of the pocket during repeated procedures, which promotes latent infection. The inoculated pathogens may survive within the pocket and become activated in the case of immunocompromise.²²^,²³ Repeated interventions weaken patients’ defense against pathogens and increase the risk of infection. One of the proposed strategies to reduce the CDI rate related to system replacement or upgrade procedures is removal of the fibrous capsule surrounding the device with electrocautery.²² Recently, Goldenberg et al²⁴ indicated that capsule debridement significantly reduced the CDI risk (with hazard ratio 0.32). Moreover, prior reports also showed that capsule removal may facilitate the penetration of antibiotics routinely administered as prophylaxis of CIED procedures and thus prevent CDI development.²⁵ The capsulectomy approach was widely used in the present study. On the other hand, Lakkireddy et al²⁶ demonstrated that capsulectomy may increase the risk of pocket hematoma. The reported frequency of pocket hematoma was 6.1% in patients with capsulectomy and 0.8% in patients without capsulectomy; however, those investigators performed follow-up evaluation without a clear standardized hematoma definition, which was a potential study limitation. In our study, we performed ultrasonographic assessment of all SPHs and their overall frequency was comparable to that in previous reports.¹⁴^,²⁷

Another proposed strategy to reduce CDI risk is application of an absorbable, antibiotic-eluting envelope as standard care to prevent CDI development.²⁸^–³⁰ The results of the recent WRAP-IT study showed a 40% reduction of CDI as compared with standard-of-care infection-prevention strategies alone.²⁹

The association between SPH and subsequent CDI occurrence has been repeatedly demonstrated in previous reports, and SPH was also one of the strongest predictor of CDI in our study.⁷^,¹⁰ In the multivariable analysis SPH increased CDI risk almost 5-fold. Interestingly, we calculated the volume of each SPH and to our knowledge, this is the first study to correlate SPH volume evaluated by ultrasound measurements with the risk of CDI. We used the ABC/2 formula previously developed for intracranial hematoma volume assessment¹⁵^,¹⁶ to calculate the SPH volume, because of the similar ellipsoid shape of SPHs. Although the limitations of such macroscopic calculations should be considered, we documented that SPH volume >200 cm³ strongly enhanced the risk of the CDI development. We performed SPH volume calculations to objectively clarify the definition of SPH and to assess the relationship between pocket bleeding complications and risk of CDI. Prior studies identified hematoma with an estimated area >100 cm² as significant, but our study showed that SPH volume >200 cm³ carried the highest CDI risk.²⁷

Several mechanisms involved in SPH-related CDI development have been suggested. One of them is tissue tension caused by the SPH leading to the wound dehiscence and further pocket contamination. Another one is necrosis of lower tissues caused by SPH mass pressure.¹⁰^,²⁷ In addition, the SPH provides a suitable environment for sustained bacterial colonization. It should be highlighted that the results of empiric antibiotic use in the case of SPH development for subsequent CDI prevention are inconclusive.⁶^,⁹^,¹⁰ Essebag et al¹⁰ reported no infection risk reduction by empiric antibiotic use, which may be caused by poor antibiotic penetration into the pocket space.¹⁰ However, some previous studies documented that lack of empiric antibiotic therapy increased the risk of the CDI.²⁰ In present study, all patients with SPH received empiric antibiotics.

Study Limitations

This study is burdened by several limitations. One is its observational nature, but that may also be considered as a strong point and reflects the real-life nature of the study. The study included a small number of patients from a single center and a study group that partially overlapped with that in the previously published PACE DRAP study.¹³ An important study limitation is the method of SPH volume estimation, which may lead to underestimation of the results. The study is also limited by the relatively short 1-year follow-up, which may be too short to reveal late complications. Finally, only CDI cases were included. Although the primary source of infection is generally attributed to pockets, it is hard to definitely exclude other entry sites and device system infection as a result of secondary seeding.

Conclusions

We identified 5 independent risk factors of CDI development: age >75 years, system upgrade procedure, procedure duration >1 h, presence of SPH and reintervention within 1 month. Our study showed that the PACE DRAP score was better able to identify patients at high risk of CDI than the PADIT score. We believe that our findings will help physicians in identifying patients at high risk of CDI. Nonetheless, further studies of larger cohorts are needed to validate the accuracy of the PACE DRAP and PADIT scores in predicting CDI.

Acknowledgments

We thank the patients and their families for participation in this study.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Financial Support

None to declare.

Disclosure Statement

No conflicts to disclose.

IRB Information

Bioethics Committee of Poznan University of Medical Sciences – reference no. 613/15.

References

1. Blomström-Lundqvist C, Traykov V, Erba PA, Burri H, Nielsen JC, Bongiorni MG, et al. European Heart Rhythm Association (EHRA) international consensus document on how to prevent, diagnose, and treat cardiac implantable electronic device infections: Endorsed by the Heart Rhythm Society (HRS), the Asia Pacific Heart Rhythm Society (APHRS), the Latin American Heart Rhythm Society (LAHRS), International Society for Cardiovascular Infectious Diseases (ISCVID) and the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Europace 2020; 22: 515–549.
2. Shariff N, Eby E, Adelstein E, Jain S, Shalaby A, Saba S, et al. Health and economic outcomes associated with use of an antimicrobial envelope as a standard of care for cardiac implantable electronic device implantation. J Cardiovasc Electrophysiol 2015; 26: 783–739.
3. Ishiguchi H, Ishikura M, Yoshida M, Imoto K, Sonoyama K, Kawabata T, et al. Incidence and risk factors for cardiac implantable electronic device infection in current clinical settings in a Japanese population: A 20-year single-center observational study. J Cardiol 2020; 76: 115–122.
4. Olsen T, Jørgensen OD, Nielsen JC, Thøgersen AM, Philbert BT, Johansen JB. Incidence of device-related infection in 97 750 patients: Clinical data from the complete Danish device-cohort (1982 – 2018). Eur Heart J 2019; 40: 1862–1869.
5. Płońska-Gościniak E, Olędzki S, Kukulski T, Kosmala W, Młynarski R, Oko-Sarnowska Z, et al. Pol-CDRIE registry: 1-year observational data on patients hospitalized due to cardiac device-related infective endocarditis in Polish referential cardiology centres. Kardiol Pol 2019; 77: 561–567.
6. Krahn AD, Longtin Y, Philippon F, Birnie DH, Manlucu J, Angaran P, et al. Prevention of Arrhythmia Device Infection Trial: The PADIT Trial. J Am Coll Cardiol 2018; 72: 3098–3109.
7. Joy PS, Kumar G, Poole JE, London B, Olshansky B. Cardiac implantable electronic device infections: Who is at greatest risk? Heart Rhythm 2017; 14: 839–845.
8. Jędrzejczyk-Patej E, Mazurek M, Kowalski O, Sokal A, Kozieł M, Adamczyk K, et al. Device-related infective endocarditis in cardiac resynchronization therapy recipients: Single center registry with over 2500 person-years follow up. Int J Cardiol 2017; 227: 18–24.
9. Birnie DH, Wang J, Alings M, Philippon F, Parkash R, Manlucu J, et al. Risk factors for infections involving cardiac implanted electronic devices. J Am Coll Cardiol 2019; 74: 2845–2854.
10. Essebag V, Verma A, Healey JS, Krahn AD, Kalfon E, Coutu B, et al. Clinically significant pocket hematoma increases long-term risk of device infection: BRUISE CONTROL INFECTION study. J Am Coll Cardiol 2016; 67: 1300–1308.
11. Uslan DZ, Gleva MJ, Warren DK, Mela T, Chung MK, Gottipaty V, et al. Cardiovascular implantable electronic device replacement infections and prevention: Results from the REPLACE Registry. Pacing Clin Electrophysiol 2012; 35: 81–87.
12. Sticherling C, Marin F, Birnie D, Boriani G, Calkins H, Dan GA, et al. Antithrombotic management in patients undergoing electrophysiological procedures: A European Heart Rhythm Association (EHRA) position document endorsed by the ESC Working Group Thrombosis, Heart Rhythm Society (HRS), and Asia Pacific Heart Rhythm Society (APHRS). Europace 2015; 17: 1197–1214.
13. Sławek-Szmyt S, Araszkiewicz A, Grygier M, Szmyt K, Seniuk W, Waśniewski M, et al. PACE DRAP: A simple score for predicting significant bleeding complications after cardiac implantable electronic device surgery. Pol Arch Intern Med 2020; 130: 206–215.
14. Birnie DH, Healey JS, Wells GA, Verma A, Tang AS, Krahn AD, et al. Pacemaker or defibrillator surgery without interruption of anticoagulation. N Engl J Med 2013; 368: 2084–2093.
15. Kwak R, Kadoya S, Suzuki T. Factors affecting the prognosis in thalamic hemorrhage. Stroke 1983; 14: 493–500.
16. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 1996; 27: 1304–1305.
17. Habib G, Lancellotti P, Antunes MJ, Bongiorni MG, Casalta JP, Del Zotti F, et al. 2015 ESC guidelines for the management of infective endocarditis: The Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur Heart J 2015; 36: 3075–3128.
18. Balla C, Brieda A, Righetto A, Vitali F, Malagù M, Cultrera R, et al. Predictors of infection after “de novo” cardiac electronic device implantation. Eur J Intern Med 2020; 77: 73–78.
19. Jędrzejczyk-Patej E, Mazurek M, Kowalski O, Sokal A, Adamczyk K, Morawski S, et al. Outcomes in patients undergoing cardiac resynchronisation therapy complicated by device-related infective endocarditis. Kardiol Pol 2018; 76: 1525–1533.
20. Polyzos KA, Konstantelias AA, Falagas ME. Risk factors for cardiac implantable electronic device infection: A systematic review and meta-analysis. Europace 2015; 17: 767–777.
21. Ząbek A, Boczar K, Dębski M, Pfitzner R, Ulman M, Holcman K, et al. Indications for transvenous lead extraction and its procedural and early outcomes in elderly patients: A single-center experience. Pol Arch Intern Med 2020; 130: 216–224.
22. Chu XM, Yu H, Sun XX, An Y, Li B, Li XB. Identification of bacteriology and risk factor analysis of asymptomatic bacterial colonization in pacemaker replacement patients. PLoS One 2015; 10: e0119232.
23. Margey R, McCann H, Blake G, Keelan E, Galvin J, Lynch M, et al. Contemporary management of and outcomes from cardiac device related infections. Europace 2010; 12: 64–70.
24. Goldenberg GR, Barsheshet A, Bishara J, Kadmon E, Omelchencko A, Strasberg B, et al. Effect of fibrotic capsule debridement during generator replacement on cardiac implantable electronic device infection risk. J Interv Card Electrophysiol 2020; 58: 113–118.
25. Nof E, Epstein LM. Complications of cardiac implants: Handling device infections. Eur Heart J 2013; 34: 229–236.
26. Lakkireddy D, Pillarisetti J, Atkins D, Biria M, Reddy M, Murray C, et al. IMpact of pocKet rEvision on the rate of InfecTion and other CompLications in patients rEquiring pocket mAnipulation for generator replacement and/or lead replacement or revisioN (MAKE IT CLEAN): A prospective randomized study. Heart Rhythm 2015; 12: 950–956.
27. Airaksinen KE, Korkeila P, Lund J, Ylitalo A, Karjalainen P, Virtanen V, et al. Safety of pacemaker and implantable cardioverter-defibrillator implantation during uninterrupted warfarin treatment: The FinPAC study. Int J Cardiol 2013; 168: 3679–3682.
28. Shariff N, Akthar T, Razak E, Segerson N, Schwartzman D. Cardiovascular implantable electronic device infections: Risk scoring and role of antibiotic envelope in prevention. Recent Adv Cardiovasc Drug Discov 2015; 10: 70–76.
29. Tarakji KG, Mittal S, Kennergren C, Corey R, Poole JE, Schloss E, et al. Antibacterial envelope to prevent cardiac implantable device infection. N Engl J Med 2019; 380: 1895–1905.
30. Koerber SM, Turagam MK, Winterfield J, Gautam S, Gold MR. Use of antibiotic envelopes to prevent cardiac implantable electronic device infections: A meta-analysis. J Cardiovasc Electrophysiol 2018; 29: 609–615.

Corresponding author

Register with J-STAGE for free!