Background
Cardiac implantable electronic devices (CIEDs) mainly include pacemaker, implantable cardiac defibrillator (ICD), cardiac resynchronization therapy (CRT). Of these, pacemakers are the most common and effective way to treat bradycardia arrhythmia (pacemaker), while ICD and CRT are effective strategies to prevent sudden death and improve heart failure, respectively. The number of CIEDs implantations have increased significantly during the past decade. There were 560,000 CIEDs implanted in the United States every year [
1,
2]. Accordingly, the number of infectious complications attributed CIEDs implantation also increased dramatically [
1,
3]. Subcutaneous pocket infection, which is mainly due to
staphylococcus aureus [
4], is the most serious infectious complications [
5,
6]. The reported incidence of these CIEDs-related infections has been ranging from 0 to 12.6% in studies [
3,
7,
8], mostly ranging from 1 to 7% [
9].
Among all the complications, pacemaker/ICD/CRT pocket infection remains a serious, and even, potentially life-threatening complication [
8]. Once pocket infection occurred, it will prolong hospital stay, increase medical cost, especially when a removal of the entire system with subsequent re-implantation is needed [
10‐
12]. Moreover, the increased incidence of infectious complication is associated with substantial elevated morbidity and mortality [
8,
13,
14].
Although there are data supporting systemic use of antibiotic for preventing infection during and after CIEDs implantations, many employ other accompanying strategies for further prevention [
2,
15]. One of the strategy is pocket irrigation with antimicrobial solutions, although there is no conclusive evidence demonstrating its benefit. Indeed, results from individual trials were conflicting and not convincing. For example, results from Lakkireddy and colleagues showing an infructuous effect of pocket irrigation while other claimed a reduced rate of infection [
16,
17]. To summary current evidence and draw a plausible conclusion, we performed a meta-analysis of available trials to re-evaluate the effectiveness of pocket irrigation with antimicrobial agents in reducing pocket infection during CIEDs implantation.
Methods
Literature search and study selection
All studies reporting the effects of pocket irrigation during CIEDs implantation published before June 2017 were identified by the comprehensive computer-based search of PubMed, Cochrane, EmBase, Web of Science, Chinese BioMedical, Global Health and BIOSIS Previews databases. The following terms were used for search: pacemaker, cardiac implantable device, VVI, DDD, implanted cardiac defibrillator, ICD, cardiac re-synchronization therapy and CRT, which were combined with pocket, irrigation, and infection. Hand searches for related articles were also performed. All the searches were conducted without language restriction. Reference lists of the retrieved articles were also reviewed to ensure to no eligible study missed.
For inclusion, procedure of the implantation had to be described properly to ensure that no operative factors leading to infection. Exclusion criteria were as follows: 1) no recording infectious rates between groups; 2) non-medication treatments of the pocket; 3) including pericardial lead implantation; and 4) sample size less than 100.
Data extraction and quality assessment
Two investigators (Kang FG and Zhang HF) extracted data independently. All the data were extracted using a standardized data-collection form. Information was recorded as follows: last name of the first author, year of publication, geographical location, study design, agents for irrigation, time from implantation to infection, duration of the follow-up period and number of participants.
The quality of enrolled studies were also assessed by the same investigators and the following elements were considered [
18]: study design, characteristics of the studied population, assessment of outcome, duration of follow-up, and statistical control for potential confounding factors. Any disagreement were resolved by a discussion. All the data were extracted from published results and written informed consent for participation was not applicable.
Statistical analysis
Meta-analysis was performed as our previous report [
19,
20]. In brief, heterogeneity of effect size across studies was quantified by the
I
2
-statistic and tested by a Cochrane
Q-test with a significance level of
P < 0.1, rather than 0.05 [
21]. Pooled effect size was estimated by Mantel-Haenszel fixed-effects model if no significant heterogeneity existed. Otherwise, the DerSimonian-Laird random-effects model was adopted. Potential publication bias was assessed by Egger’s test and Begg’s funnel plot was produced [
22].
To further investigate the effects of pocket irrigation during CIEDs implantations, subgroup analyses according to study designs, classes of medications, time from implantation to infection, and geographical locations were performed. Meta-regression using restricted maximum likelihood estimation method was adopted to explore sources of heterogeneity and to confirm results from subgroup analyses [
23].
P-values were adjusted with the Monte Carlo method (permute = 100) if more than 3 co-variables enrolled in the regression model to reduce type I error [
24]. Finally, a sensitivity analysis, which investigated the influence of a single study on the overall risk estimated by omitting one study in each turn, was used to test the stability of the pooled results.
The study was performed in accordance with the PRISMA statement [
25]. All analyses were performed by using STATA version 12.0 and graph of quality assessment was produced by Revman 5.3.
Discussion
Pocket infection is the most serious complication of CIEDs implantation and current evidence on the effects of pocket irrigation is inconsistent. To summary available information in this field, the meta-analysis of 10 studies involving 5467 participants supported a protective effect of antibiotics pocket irrigation, irrespective of antibiotics classes. Compared with the use of saline, the incidence of pocket infection was reduced by about 59% with antibiotic irrigation.
Implantations of CIEDs have increased over the past decades, despite implantation technique improvement, the infection rate has increased dramatically [
2,
8]. Pocket infection may be due to the following causes [
34]: (1) Pacing leads and pulse generators are exposed to the air before implantation, which may cause airborne pathogens; (2) To the human body, the implant is a metallic foreign body which has a rough surface, making it prone to bacteria growth; (3) Operation duration, which is greatly affected by the individual condition of the patient and the technique of the operator; (4) The ambient air quality in the operating room, which may not be sufficient to reach the real sterile in some time; (5) Patients receiving CIEDs implantation are mostly aged ones, who could have various co-existing diseases, making them more prone to be infected. It is obvious that most of the reasons could not be completely avoided and strategies to prevent pocket infect becoming crucial.
Besides systematic application of antibiotics, little effective strategy is known to reduce pocket infection. Therefore, screening and confirming positive methods to reduce pocket infection is of great importance. Pocket irrigation may be a promising strategy for reducing pocket infection but debate on this issue is persistent during the decade [
7,
35]. Indeed, pocket infection, despite increasing, is not with high prevalence, which result in a need of large sample size to give a statistical significance. It may be difficult for a single study to reach such a large sample size. Because of the inconsistent results of the trials, current guideline did not give a recommendation on pocket irrigation to reduce pocket infection [
36]. Therefore, pooled analysis of existing information in this field becoming important and emergent. The current study pooled all the data on pocket irrigation and showed a benefit of reducing infection, irrespective of class of antibiotics, which robustly supported that the use of antibiotics for pocket irrigation should be encouraged. Indeed, an survey of 2092 electro-physiologists in more than fifty countries believed that the use of anti-microbial agents for pocket irrigation could reduce infection of CIEDs implantation [
35]. Thus, the current study provide conclusive evidence supporting this view point.
The same important is to identify the most effective drugs for irrigation. As known, the most common bacteria causing pocket infection is
staphylococcus aureus [
12,
36], which accounts for about 60 to 80% of the pathogens in infection [
6]. For the treatment of
staphylococcus aureus, cefazolin and gentamicin were recognized as the first line therapy while others were believed as the second line [
6]. As indicated in the present study, the second line therapy for
staphylococcus aureus was more effetive despite the first line therapy also conferred protective effects. Therefore, when choosing the medication for pocket irrigation for whom are proning to be infected, drugs of the second line therapy may be considered first. Besides, drug resistance of the bacteria in their own center should also be taken into consideration. For example, in south of China, cefazolin is the most frequently used while the third generation of cephalosporin and lincosamide were adopted in north of China and the USA respectively [
31,
33].
Some limitations of the current study should be noted. Firstly, despite subgroup analysis of RCTs/non-RCTs also showed an ineffectiveness of retrospective studies, meta-regression did not support study design as the variable for variance. Of note, one of the retrospective study using non-antibiotics, which was shown to be the source of the heterogeneity. Therefore, it is unknown whether the retrospective study did not support the adoption of pocket irrigation in nature, or like co-variable of region, which is actually masked by the use of non-antibiotics in the retrospective study. Secondly, in-sufficient data restricted further analysis according to different kinds of CIEDs (VVI/DDD/ICD/CRT/CRT-D) implantations, which might also be an important issue since patients receiving ICD/CRT/CRT-D were largely heart failure or ventricular arrhythmia survivors, who may be much more complicated and have distinct properties concerning with pocket infection.
Conclusion
The current study demonstrated significantly protective effect on preventing pacemaker pocket infection with antibiotics pocket irrigation, irrespectively of classes of antibiotics, which robustly supported regular use of antibiotics for pocket irrigation during CIEDs procedure. As well as a need for large sample size confirming this conclusion in prospective studies with well control of bias, there is also an urgent need for screening of most effective medication, which together contribute to the reduction of pocket infection.