Figures
Abstract
Background
The use of cardiac implantable electronic devices (CIED) has risen steadily, yet the rate of cardiac device infections (CDI) has disproportionately increased. Amongst all cardiac device infections, the pocket infection is the most challenging diagnosis. Therefore, we aimed to improve diagnosis of such pocket infection by identifying relevant biomarkers.
Methods
We enrolled 25 consecutive patients with invasively and microbiologically confirmed pocket infection. None of the patients had any confounding conditions. Pre-operative levels of 14 biomarkers were compared in infected and control (n = 50) patients. Our selected biomarkers included white blood cell count (WBC), C-reactive protein (CRP), procalcitonin (PCT), lipopolysaccharide binding protein, high-sensitivity C-reactive protein (HS-CRP), polymorphonuclear-elastase, presepsin, various interleukins, tumor necrosis factor α (TNF-α), and granulocyte macrophage colony-stimulating factor (GM-CSF).
Results
Of the 25 patients with isolated pocket infection (70±13years, 76% male, 40% ICDs), none presented with leukocytosis. In contrast, they had higher serum levels of HS-CRP (p = 0.019) and PCT (p = 0.010) than control patients. Median PCT-level was 0.06 ng/mL (IQR 0.03–0.07 ng/mL) in the study group versus 0.03 ng/mL (IQR 0.02–0.04 ng/mL) in controls. An optimized PCT cut-off value of 0.05 ng/mL suggests pocket infection with a sensitivity of 60% and specificity of 82%. In addition TNF-α- and GM-CSF-levels were lower in the study group. Other biomarkers did not differ between groups.
Citation: Lennerz C, Vrazic H, Haller B, Braun S, Petzold T, Ott I, et al. (2017) Biomarker-based diagnosis of pacemaker and implantable cardioverter defibrillator pocket infections: A prospective, multicentre, case-control evaluation. PLoS ONE 12(3): e0172384. https://doi.org/10.1371/journal.pone.0172384
Editor: Yoshihiro Fukumoto, Kurume University School of Medicine, JAPAN
Received: September 25, 2016; Accepted: February 3, 2017; Published: March 6, 2017
Copyright: © 2017 Lennerz et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: The German Heart Centre received an unearmarked grant of 10,000 EUR by Biotronik GmbH to support research activities in the broad field of cardiac rhythm management and cardiac devices. The funder had no role in selection of the research topic, study design, datacollection and analysis, decision to publish, or preparation of the manuscript. This research grant was used to purchase the required reagent kits for processing the blood samples and analysing the biomarkers.
Competing interests: The German Heart Centre received an unearmarked grant of 10,000 EUR by Biotronik GmbH to support research activities in the broad field of cardiac rhythm management and cardiac devices. This research grant was used to purchase the required reagent kits for processing the blood samples and analysing the biomarkers. In addition, CL and CK disclose their general relation with several pacemaker and ICD manufacturers including Biotronik GmbH. CL received lecture fees unrelated to the subject matter of this paper from Biotronik GmbH. CL also received travel support from Biotronik GmbH to present the results of this study at the German Society of Cardiology Congress (Herbsttagung) and at the ESC Congress 2014 in Barcelona. CK also received lecture honoraria and travel support from Biotronik GmbH unrelated to the subject matter of this paper. CK has a consulting relationship with many ICD device manufacturers, including Biotronik, Boston Scientific, and Sorin. He also served as a principal investigator on clinical studies sponsored by several manufacturers (Biotronik, Boston Scientific, Medtronic, St. Jude Medical, Sorin). This declaration of research association with Biotronik includes the current study, for which CK invested the above mentioned research grant awarded to the author´s institution. There are no patents, products in development or marketed products to declare. This does not alter our adherence to all the PLOS ONE policies on sharing data and materials.
Introduction
Implantable pacemakers (PM) and cardioverter defibrillators (ICD) are standard treatment for bradyarrhythmias, to ameliorate heart failure, and to protect against sudden cardiac death [1]. Worldwide implantation rates of cardiac implantable electronic devices (CIED) are estimated at 1,250,000 pacemakers and 410,000 cardioverter defibrillators per year, with an annual increase of roughly 5% [2–4]. The recent increase in device utilisation was driven by the needs of an aging population coupled with an expansion of CIED functions and also their indications [5–9]. Nevertheless, increased device use has also increased the incidence of complications. Moreover, these infections have increased disproportionately versus the rate of implantation [2, 10, 11]. This trend in cardiac device infection (CDI) burden is also attributed to the aging population together with an increased prevalence of co-morbidities in current device recipients [2, 5]. The typical estimated annual infection rate is between 1% and 2%; although published rates range from 0.5% to 8% [2, 12–19]. CDI are, not surprisingly, associated with increased costs, significant morbidity, and higher mortality [13, 20–23].
CDI can be classified into three main types; [1] isolated pocket infection, [2] bloodstream infection, or [3] cardiac device related infective endocarditis (CDRIE). Isolated pocket infection is the most frequent, accounting for more than half of CDI [24, 25]. Early diagnosis and subsequent complete device and lead removal, combined with antibiotic treatment, are important to avoid progression to sepsis or endocarditis [5, 16, 26, 27]. However, diagnosing a pocket infection can be challenging because many patients present with few or very mild symptoms or even without any obvious signs of local infection [24, 28–30].
Such frequent lack of clear-cut symptoms of infection place the onus on clinical judgment [28]. Conventional inflammation-related biomarkers, such as white blood count or erythrocyte sedimentation rate, are known to exhibit low sensitivity to cardiac device pocket infections. Hence, they rarely influence diagnostic decisions [24, 31–32]. Therefore, we aimed to evaluate additional inflammatory biomarkers to determine if they enhance diagnosis of cardiac device pocket infection.
Methods
The DIRT (Device associated infections–role of new diagnostic tools) study is a prospective, multicentre, case-control evaluation of inflammatory biomarkers in a cohort of CIED recipients with and without pocket infections. Patients were recruited from five centres in Germany, Croatia, and Italy. From August 2011 to October 2012, consecutive patients with suspected cardiac device pocket infections and control patients at the time of pulse generator exchange or lead revision were invited to participate in the study. The study was approved by the ethical review board of the Technische Universität München, Munich, and conducted according to the principles of the Declaration of Helsinki. The study was registered at clinicaltrials.gov (identifier, NCT01619267). All patients provided written informed consent and 50 mL of peripheral blood were drawn for biomarker analysis.
Inclusion and exclusion criteria
The inclusion criterion for the evaluation of inflammatory biomarkers was the presence of a cardiac device pocket infection (as described below). Controls were recruited from patients presenting for a pulse generator exchange or lead revision (unrelated to infections) at the same centres during the same period.
Exclusion criteria (for both groups) were the presence of a bloodstream infection or an infective endocarditis; attributed according to modified Duke criteria [33, 34]. Patients were excluded if they presented with current antibiotic treatment, or other concomitant infectious, or inflammatory disease. Additional exclusion criteria were circumstances that could influence inflammatory biomarker levels; e.g. recent trauma, surgery, or burns, patients with active or recent (within two years) malignancy, patients receiving systemic steroid therapy, and patients on high-flux renal dialysis. Minors or adults under guardianship were excluded.
Diagnostic assessment
Fifty-three patients with cardiac device infection were screened and classified (Fig 1). All patients underwent a baseline assessment, which included; the medical history, the indication for device therapy and device related procedures, physical examination, and basic laboratory tests including white blood count and C-reactive protein (CRP). In all patients with suspected CDI, at least three sets of blood cultures, prior to initiation of antibiotic therapy, were drawn in order to assess potential blood stream infection or infective endocarditis. Additionally, in all these patients, transthoracic and transoesophageal echocardiography was performed to detect valvular vegetation or newly developed valvular insufficiency (Fig 2).
An isolated generator pocket infection was assumed (in the absence of a bloodstream infection or infective endocarditis) if at least three out of the following seven local signs of inflammation or infection were present: erythema, pain, warmth, swelling, induration, tenderness, or fluctuation. Furthermore, hardware protrusion through the skin or pus discharge at the pocket site (either spontaneous or expressed upon palpation of the site) was considered conclusive evidence of pocket infection. In all patients with suspected pocket infection, the diagnosis was required to be confirmed by surgical exploration of the generator pocket site showing purulent or inflammatory changes. In cases of invasively confirmed CDI, we removed all hardware using a transvenous lead extraction approach. In all patients with confirmed CDI, cultures from the device pocket and from the leads were taken for microbiological analysis. If exploration of the pocket failed to reveal a CDI, or if it had not been performed, the patient was ineligible for the study.
After this assessment, patients were classified to have either an infection limited to the generator pocket–representing the target population–or to have an infection with systemic components (i.e., bloodstream infection or definite and possible infectious endocarditis according to the modified Duke criteria) and therefore excluded. [12, 33–34]
Biomarker selection and analysis
In addition to basic laboratory tests of white blood count (WBC) and serum C-reactive protein (CRP), we assessed 12 other biomarkers that could potentially support the diagnosis of cardiac device pocket infection: procalcitonin (PCT), high-sensitivity CRP (HS-CRP), lipopolysaccharide binding protein (LBP), presepsin, polymorphonuclear-elastase (PMN-E), interleukins (IL)-1ß, -6, -8, -10, -23, tumor necrosis factor α (TNFα), and granulocyte macrophage colony-stimulating factor (GM-CSF). The biomarker selection was primarily based on a systematic literature review, searching PubMed with terms related to cardiac device infections and bacterial endocarditis. The query resulted in the identification of 10 biomarkers as promising candidates to detect infection: WBC, CRP, HS-CRP, PCT, LBP, IL-1β, IL-6, IL8, TNF-α and PMN-Elastase (Fig 3) [35–48]. Given the leading causative agents for device infection (bacteria in 80–90% of cases; predominantly coagulase negative staphylococci and staphylococci aureus in 60–70% of cases), and taking into account the inflammatory nature of the process and the specific inflammatory cascades involved, we solicited expert opinion; basic science researchers in immunology or infectious disease and laboratory medicine specialists for their advice on the selection of additional biomarkers for the detection of chronic bacterial infections. Presepsin, IL-10. IL-23 and GM-CSF were the markers suggested by these experts as worthy of evaluation (subsequently substantiated by literature review; Fig 3) [49–52]. The aim of testing numerous biomarkers was to cover all inflammatory process cascades and to identify candidates for subsequent follow-up in a larger study.
After informed consent, and prior to any antibiotic treatment or any invasive procedures, 50 mL of blood were drawn from study participants. The blood samples were divided between heparin, EDTA, citrate, and plasma tubes–according to the assay method–centrifuged, and stored at -70°C. Analysis of biomarker concentrations was performed in a core-lab at the German Heart Centre Munich using commercially available assays (Table 1). All analyses were performed according to the manufacturers´ instructions.
Endpoint
The study endpoint was to assess the accuracy of 14 biomarkers in diagnosis of cardiac device infection limited to the generator pocket.
Statistics
The DIRT study was designed as an explorative evaluation of the diagnostic accuracy of biomarkers associated with cardiac device pocket infections. For this, patients were prospectively and consecutively included until the target population—patients with isolated pocket infection naïve to antibiotic pre-treatment—reached 25 and the control group reached 50. Because there was no current data to indicate the potential efficacy of these biomarkers, our study was intended as a preliminary evaluation and hence no pre-investigation power-analysis was performed. Nevertheless, our predefined sample sizes were considered large enough to provide sufficient discrimination.
Categorical data are presented as absolute and relative frequencies, continuous data as mean ± standard deviation or as median with interquartile range (IQR). Comparisons were performed using either chi-square or Fisher exact tests for categorical variables as appropriate. Continuous variables were analysed using a two-sample t-test if normally distributed. Otherwise, the Mann Whitney U-test was used. Initially, for inflammatory biomarkers with established cut-off values, the diagnostic accuracy for isolated pocket infection was described by values of sensitivity and specificity. Subsequently, receiver operating characteristic (ROC) curves were drawn for all biomarkers. The area under the ROC curve (AUC) with 95% confidence intervals was then calculated. Finally, the optimal cut-off value of each biomarker (i.e., the maximized sum of sensitivity and specificity; Youden index) was derived. Statistical analyses were performed using SPSS V.21.0 (IBM Corporation, Armonk, USA).
Results
Study samples
According to the modified Duke criteria, cardiac device related infective endocarditis (CDRIE) was diagnosed in five patients, bloodstream infection was confirmed in three, and pocket infection was suspected in 45. After application of the inclusion and exclusion criteria, we analyzed blood samples from 25 patients with invasively confirmed pocket infections (Fig 1).
Baseline characteristics
Demographic and baseline characteristics are summarized in Table 2. Generally, these did not differ between groups. However, there was weak evidence to suggest that infected patients were more likely to have ischemic cardiomyopathy and to have undergone more CIED procedures more recently. Swab- or lead-cultures were available in 24 out of 25 of patients and were positive in 20 out of 24 (83%) individuals. Pathogens in patients with isolated pocket infection without antibiotic pretreatment included: Staphylococcus epidermidis (n = 12, 50%), Staphylococcus capitis (n = 4, 17%), Staphylococcus aureus (n = 2, 8%), Staphylococcus haemolyticus (n = 1, 4%), Pseudomonas aeruginosa (n = 1, 4%); no specific pathogen was identified in 4 (17%) patients.
Biomarker evaluation using established cut-offs
None of the routinely used laboratory biomarkers (WBC, CRP, HS-CRP, and PCT) were associated with the presence of pocket infection when established reference values were used. Notably, none of the participants presented with leukocytosis and the serum procalcitonin (PCT)-level never exceeded the upper reference limit of 0.5 ng/mL used in the diagnosis of endocarditis or sepsis. CRP and HS-CRP levels were more often elevated in the study group versus the controls. The difference was more pronounced for HS-CRP; however, the evidence was weak.
In contrast, the less frequently used biomarkers provided stronger evidence of inter-group differences. Presepsin levels above the established cut-off were more prevalent in patients with pocket infections than in healthy controls; the sensitivity and specificity for pocket infection diagnosis was moderate. GM-CSF was found to be significantly more frequently elevated above the reference value in controls than in patients with pocket infections, however, sensitivity and specificity was low. Table 3 provides the details for each of the investigated biomarkers.
Biomarker evaluation using absolute concentrations
For each biomarker, the median concentrations and their IQRs and the AUC for each ROC are shown in Fig 4. White blood counts did not differ between groups. However, infected patients had statistically significantly higher serum levels of C-reactive protein (CRP, 2.7mg/L vs. 1.6mg/L, 95% CI 0.53 to 0.79, p = 0.028), high-sensitivity CRP (HS-CRP, 3.1mg/L vs. 1.7mg/L, 95% CI 0.53 to 0.90, p = 0.019), and procalcitonin (PCT, 0.06ng/mL vs. 0.03ng/mL, 95% CI 0.55 to 0.82, p = 0.01) than controls. In contrast, GM-CSF- (0.12pg/mL vs. 0.59pg/mL, 95% CI 0.56 to 0.81, p = 0.01) and TNF-α-levels (0pg/mL vs. 1.1pg/mL, 95% CI 0.77 to 0.97, p<0.01) were lower in infected patients versus controls. The concentrations of presepsin, LBP, PMN-Elastase, and the tested interleukins (IL-1β, IL-6, IL-8, IL-10, IL-23) did not differ between groups. The five biomarkers (CRP, HS-CRP, PCT, TNF-α and GM-CSF) with the largest apparent potential to differentiate pocket infections are illustrated in Fig 5 as ROC curves.
In addition, optimized cut-off values with maximized sensitivity and specificity were obtained from ROC analysis applying the Youden Index. For the biomarkers established in routine diagnostics (CRP, HS-CRP, PCT), the optimized cut-offs resulted in moderate diagnostic power to discriminate between patients with isolated cardiac device pocket infections and healthy controls. Of these, PCT yielded the best combination of sensitivity and specificity. The optimized PCT threshold of 0.05 ng/mL, which is above the 95th percentile of that found in a healthy population, but 10-times lower than the established cut-off for septic conditions, achieved a sensitivity of 60% and specificity of 82% (Table 4). Two of the additional biomarkers tested (TNF-α and GM-CSF), yielded comparable, or somewhat higher, sensitivity and specificity levels with optimized cut-offs. For both markers, low serum concentrations were associated with the presence of infection (Table 4).
Discussion
Diagnosis of pocket infections for implanted cardiac devices is challenging and requires considerable experience [35, 48]. Experience is needed first to initiate necessary treatment early enough to avoid development of sepsis or endocarditis. Second, experience is required to avoid unnecessary pocket exploration, with its associated risks, in uninfected patients even when the pocket appears suspicious. Given this spectrum of necessary to unnecessary intervention, biomarkers could aid diagnosis and thereby assist the decision-making process.
In the DIRT study, we provide the first ever assessment of the ability of biomarkers associated with inflammatory processes to differentiate between patients with and without cardiac device pocket infections. Data on such differentiation is scarce and limited to measurement of WBC- and CRP-levels. However, in the majority of patients with cardiac device pocket infection, WBC and CRP levels are normal and therefore provide minimal diagnostic value [24, 32]. These prior observations are confirmed in our study and further underline the unsuitability of non-elevated WBC- and CRP-levels to rule out CDI.
The DIRT study was not restricted to these standard infection parameters, but also assessed 12 additional biomarkers. The group of chosen biomarkers are not specific for the germs causing pocket infections. However, because they cover the different cascades of the inflammatory processes, they represent promising candidates to detect isolated pocket infections. (Fig 3) We used biomarkers with proven responsiveness to bacteria because of the predominately bacterial character of CIED-infection. However, we also included a broad set of interleukins (involved in the specific and unspecific cascade) in order to detect other independent inflammatory processes. This concept of detection of inflammatory reactions is also the same as that underlying the recommended use of F-FDG-PET/CT or SPECT/CT in suspected endocarditis.
When we applied standard cut-offs, increased serum levels of presepsin, and decreased levels GM-CSF, were associated with cardiac device pocket infections. However, the respective sensitivity and specificity provided was, at best, moderate. The finding of increased presepsin serum levels in patients with cardiac device pocket infections appears intuitively reasonable; but, we cannot explain why GM-CSF expression was attenuated.
When we compared the absolute concentration of the biomarkers, we found some evidence of associations for CRP, HS-CRP, PCT, TNF-α and GM-CSF. Increased levels of CRP, HS-CRP and PCT were associated with infections, as were decreased values of TNF-α and GM-CSF. We subsequently used these associations to derive optimized cut-off values and thereby improve the sensitivity and specificity of the tests. However, some of these markers may fail to provide effective and valid screening. For example, CRP concentrations are commonly used to support diagnosis of infective endocarditis and to monitor patient response to therapy [36]. The concentrations tend to be highest in acute S. aureus infections [37], a pathogen also frequently found in cardiac device pocket infections. Application of the optimized reference value to 2.1 mg/L instead of standard 5.0 mg/L increased sensitivity and specificity to 64% and 62%, respectively. However, CRP is not specific for bacterial infections and can also be elevated in viral infections and after surgery or trauma. We excluded these conditions and so evaluation in a less selected patient population could be expected to diminish the apparent differentiating ability. Therefore, CRP does not appear an ideal parameter to detect infection. Similar arguments may also apply to HS-CRP [38]. Furthermore, in terms of sensitivity and specificity, HS-CRP did not provide any additional advantage over the use of CRP.
In contrast, procalcitonin (PCT) serum levels in suspected cardiac device pocket infection may differentiate between healthy and infected patients. Using the optimized cut-off reference value of 0.05 ng/mL (one tenth the standard cut-off value), sensitivity and specificity were 60% and 82%, respectively. This device infection specific PCT cut-off value of >0.05ng/mL is above the normal value in a healthy population. A study of 492 samples (performed with the Elecsys BRAHMS PCT assay) revealed a normal value of 0.046 ng/mL (representing the 95th percentile) [53]. Thus, patients with a PCT value of >0.05ng/mL have less than a 5% chance of a false-positive result, i.e. being a healthy subject.
PCT is known to be an accurate marker for systemic bacterial infection (independent of the pathogen) and, when compared to CRP, it is less prone to influence by viral infections, surgery, or trauma [39]. PCT has also gained importance in the diagnosis and monitoring of infective endocarditis; a differential diagnosis to isolated cardiac device pocket infection. Although PCT levels observed in infective endocarditis differ between assays and also vary according to the duration of infection, they are usually higher than those observed in our cohort of infected patients [35, 40–41]. In our experience, levels of PCT >0.5 ng/mL are typically associated with systemic infection (bloodstream and infectious endocarditis). In order to stimulate future research, we need to prove our pocket-infection-specific cut-off value prospectively. We also need to establish an upper limit that would be suggestive of systemic or septic conditions.
Relatively low–but elevated–serum concentrations of PCT in cardiac device pocket infections may be explained by the localized nature of the infection. In strictly localized infection, there is pronounced increase in PCT levels only if the infection involves surrounding tissues or becomes systemic [42]. Thus, the use of a PCT assay with a low detection limit (in our study 0.02 ng/mL) may be required.
One potential disadvantage of using PCT serum concentration is its rapid decrease in successfully treated patients [42, 43]. Therefore, ideally, blood samples should be drawn before the initiation of antibiotic treatment (as was the case in our study).
The optimized cut-offs for infection diagnosis for GM-CSF and TNF-α produced higher sensitivities and specificities versus CRP, HS-CRP, and PCT. Low serum concentrations of GM-CSF and TNF-α correlated with pocket infection. This finding appears counter-intuitive because, in general, increased expression of these regulators would be expected to occur as part of the inflammatory response. However, these low levels could represent an expression of the entity’s pathogenesis; i.e., a low expression may indicate a poor immune response which thereby facilitates infection. We acknowledge that these explanations are speculative; however, they are hypothesis-generating concepts to stimulate future research. In order to know if and how a cytokine is deregulated in a certain condition, we need to compare normal physiological values with those expressed in the pathology of interest. Curiously, prior studies failed to demonstrate elevated TNF-α-levels in patients with systemic infective endocarditis [44, 45]. The authors speculated that this occurred because of down-regulation of immune cells during persistent stimulation or because of limited ability to stimulate specific cytokines [41].
Another crucial area of research for all relevant biomarkers is to identify the date of applicability after index surgery. From studies analysing WBC count and (18)F-FDG uptake after device implantation/revision, we know the post-operative inflammation process persists for 4–8 weeks after device implantation. For example, after device implantation (<60 days), 10–15% of all patient exhibited a >50% increase in WBC count; a modest WBC count increase of 18 ± 30% was observed for the entire cohort [54]. In addition, Sarrazin et al. reported residual post-operative inflammation present 4 to 8 weeks after surgery [55]. Moreover, the current ESC guidelines on infectious endocarditis recommend a PET/CT scan for infective endocarditis only if the implantation (prosthetic) took place at least three months earlier [56]. Therefore, a conservative approach would be not to use such biomarkers until at least eight weeks after implantation. However, the current use of PCT for stewardship of antibiotic therapy duration in patients after surgery, or with community-acquired pneumonia, may provide a compelling argument that PCT levels respond quickly to changes in inflammatory and infectious conditions. [57, 58]. Also, an in vivo half-life of 22–25 hours provides a rationale for a shorter prohibited period. Nevertheless, we currently lack strong evidence to support the reliability of any biomarker immediately after surgery. However, most patients with suspected pocket infection are admitted several months, rather than weeks, after surgery.
The diagnosis of isolated pocket infection will continue to require a critical clinical awareness, careful patient history assessment, precise physical examination, and a basic workup (i.e. blood cultures, transthoracic and transoesophageal echocardiography). Although one should not draw conclusions from WBC-levels within the normal range, some laboratory parameters (e.g. PCT, CRP, HS-CRP) may provide additional information in the diagnostic decision process. Assessment of PCT has become a routine laboratory parameter used in our evaluation of patients with suspected CIED infection. In cases of possible pocket infection, when the Duke Criteria are not met and local signs are unconvincing (less than three present), we assign a PCT measurement above the cut-off value of 0.05 ng/mL a weight equal to that of a local sign. However, this diagnostic strategy needs to be assessed in a prospective study.
Limitations
The present study was designed as a pilot evaluation to identify biomarkers worthy of investigation in the diagnosis of cardiac device pocket infections. Therefore, the number of patients included was small. Nonetheless, we assessed a wide variety of biomarkers. CRP, HS-CRP, Procalcitonin, TNF-α, and GM-CSF provided the best discrimination of CDI. However, the specific values used for detection of cardiac device-associated infections differed from standard cut-offs. Our revision of these standard values should be assessed in a prospective study.
Patients with co-morbidities that might interfere with biomarker levels in CDI were excluded. Therefore, our results may some limits to their generalizability. Specifically, our cut-off values may not apply to patients on renal dialysis, patients with altered immune response (e.g., patients on steroids), or after surgery/trauma, or those with concomitant malignant diseases. Furthermore, patients with antibiotic pre-treatment were also excluded because this could influence the results.
Conclusion
Diagnosis of cardiac device pocket infection remains primarily based on the judgment of experienced physicians. CRP, HS-CRP, and procalcitonin with specific cut-offs for cardiac device infections may provide objective evidence to assist with diagnosis. In contrast, white blood count, lipopolysaccharide binding protein, presepsin, polymorphonuclear-elastase, and interleukins-1ß, -6, -8, -10, -23, do not appear to provide sufficient discrimination to aid diagnosis. The role of depressed levels of tumor necrosis factor α and granulocyte macrophage colony-stimulating factor in cardiac device pocket infections warrants further investigation.
Supporting information
S1 Trend Checklist. TREND statement checklist.
https://doi.org/10.1371/journal.pone.0172384.s001
(PDF)
Acknowledgments
We thank Herribert Pavaci (Medizinische Klinik I, Krankenhaus Landshut-Achdorf, Landshut, Germany) and Daniele Porcelli (Department of Cardiology, Ospedale Fatebenefratelli San Pietro, Rome, Italy) for recruiting participants and providing blood samples. In addition we thank Christian Grebmer, Felix Bourier, Verena Semmler, Gesa von Olshausen for recruiting participants and Tilko Reents, Amir Brkic for clinical processing of the participants and providing intraoperative data (GvO from the I. Medizinische Klinik, Klinikum rechts der Isar, Faculty of Medicine, Technische Universität München, Munich, Germany, all others work at Klinik für Herz- und Kreislauferkrankungen, Abteilung für Elektrophysiologie, Deutsches Herzzentrum München, Technische Universität München, Munich, Germany)
Author Contributions
- Conceptualization: CL SB IO CK.
- Data curation: BH TP HV.
- Formal analysis: BH SB AL CK.
- Funding acquisition: CL CK.
- Investigation: SB AL HV.
- Methodology: CL SB IO CK.
- Project administration: CL CK.
- Resources: SB IO CK ID.
- Software: BH SB.
- Supervision: SB IO CK ID.
- Validation: CL TP CB CK.
- Visualization: CL TP BP.
- Writing – original draft: CL PB CK.
- Writing – review & editing: CL PB TP CK JM PW ID.
References
- 1. Brignole M, Auricchio A, Baron-Esquivias G, Bordachar P, Boriani G, Breithardt OA, et al. 2013 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy: the Task Force on cardiac pacing and resynchronization therapy of the European Society of Cardiology (ESC). Developed in collaboration with the European Heart Rhythm Association (EHRA). Eur Heart J. 2013;34:2281–2329. pmid:23801822
- 2. Greenspon AJ, Patel JD, Lau E, Ochoa JA, Frisch DR, Ho RT, et al. 16-year trends in the infection burden for pacemakers and implantable cardioverter-defibrillators in the United States 1993 to 2008. J Am Coll Cardiol. 2011;58:1001–1006. pmid:21867833
- 3. Birnie DH, Healey JS, Wells GA, Verma A, Tang AS, Krahn AD, et al. BRUISE CONTROL Investigators. Pacemaker or defibrillator surgery without interruption of anticoagulation. N Engl J Med. 2013;368:2084–2093. pmid:23659733
- 4. Uslan DZ, Sohail MR, St Sauver JL, Friedman PA, Hayes DL, Stoner SM, et al. Permanent pacemaker and implantable cardioverter defibrillator infection: a population-based study. Arch Intern Med. 2007;167:669–675. pmid:17420425
- 5. Baddour LM, Epstein AE, Knight BP, Levison ME, Lockhart PB, Erickson CC, et al. Update on cardiovascular implantable electronic device infections and their management: a scientific statement from the American Heart Association. Circulation. 2010;121:458–477. pmid:20048212
- 6. Moss AJ, Zareba W, Hall WJ, et al., for the Multicenter Automatic Defibrillator Implantation Trial II Investigators. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med 2002;346:877–883. pmid:11907286
- 7. Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R, et al. Sudden cardiac death in heart failure trial (SCD-HeFT). Amiodarone or an implantable defibrillator for congestive heart failure. N Engl J Med 2005;352:225–237. pmid:15659722
- 8. Myerburg RJ. Implantable cardioverter-defibrillators after myocardial infarction. N Engl J Med 2008;359:2245–2253. pmid:19020326
- 9. Mond HG, Proclemer A. The 11th world survey of cardiac pacing and implantable cardioverter-defibrillators: calendar year 2009—a World Society of Arrhythmia's project. Pacing Clin Electrophysiol. 2011 Aug;34(8):1013–27. pmid:21707667
- 10. Voigt A, Shalaby A, Saba S. Rising rates of cardiac rhythm management device infections in the United States: 1996 through 2003. J Am Coll Cardiol. 2006;48;590–591. pmid:16875991
- 11. Cabell CH, Heidenreich PA, Chu VH, Moore CM, Stryjewski ME, Corey GR, et al. Increasing rates of cardiac device infections among Medicare Am Heart J. 2004;147:582–586. pmid:15077071
- 12. Ohlow MA, Lauer B, Buchter B, Schreiber M, Geller JC, Pocket related complications in 163 patients receiving anticoagulation or dual antiplatelet therapy: D-Stat Hemostat™ versus standard of care. Int J Cardiol. 2012;159:177–80 pmid:21376404
- 13. Prutkin JM, Reynolds MR, Bao H, Curtis JP, Al-Khatib SM, Aggarwal S, et al. Rates of and factors associated with infection in 200 909 Medicare implantable cardioverter-defibrillator implants: results from the National Cardiovascular Data Registry. Circulation. 2014;130:1037–1043. pmid:25081281
- 14. Mittal S, Shaw RE, Michel K, Palekar R, Arshad A, Musat D, et al. Cardiac implantable electronic device infections: incidence, risk factors, and the effect of the AigisRx antibacterial envelope. Heart Rhythm. 2014;11:595–601. pmid:24333543
- 15. Klug D, Balde M, Pavin D, Hidden-Lucet F, Clementy J, Sadoul N, et al. PEOPLE Study Group. Risk factors related to infections of implanted pacemakers and cardioverter-defibrillators: results of a large prospective study. Circulation. 2007;116:1349–1355 pmid:17724263
- 16. Wilkoff BL, Love CJ, Byrd CL, Bongiorni MG, Carrillo RG, Crossley GH 3rd, et al. Transvenous lead extraction: Heart Rhythm Society expert consensus on facilities, training, indications, and patient management: this document was endorsed by the American Heart Association (AHA). Heart Rhythm 2009;6:1085–1104 pmid:19560098
- 17. Johansen JB1, Jørgensen OD, Møller M, 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;32:991–998. pmid:21252172
- 18. Kirkfeldt RE1, Johansen JB, Nohr EA, Jørgensen OD, Nielsen JC. Complications after cardiac implantable electronic device implantations: an analysis of a complete, nationwide cohort in Denmark. Eur Heart J. 2014;35:1186–1194. pmid:24347317
- 19. Romeyer-Bouchard C1, Da Costa A, Dauphinot V, Messier M, Bisch L, Samuel B, Lafond P, et al. Prevalence and risk factors related to infections of cardiac resynchronization therapy devices. Eur Heart J. 2010;31:203–210. pmid:19875388
- 20. Le KY, Sohail MR, Friedman PA, Uslan DZ, Cha SS, Hayes DL, et al. Impact of timing of device removal on mortality in patients with cardiovascular implantable electronic device infections. Heart Rhythm. 2011;8:1678–1685. pmid:21699855
- 21. 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;171:1821–1828. pmid:21911623
- 22. Margey R1, 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. pmid:19910314
- 23. Tarakji KG, Chan EJ, Cantillon DJ, Doonan AL, Hu T, Schmitt S, et al. Cardiac implantable electronic device infections: presentation, management, and patient outcomes. Heart Rhythm. 2010;7:1043–1047. pmid:20470904
- 24. Sohail MR, Uslan DZ, Khan AH, Friedman PA, Hayes DL, Wilson WR, et al. Management and outcome of permanent pacemaker and implantable cardioverter-defibrillator infections. J Am Coll Cardiol. 2007;49:1851–1859. pmid:17481444
- 25. Tarakji KG, Wazni OM, Harb S, Hsu A, Saliba W, Wilkoff BL. Risk factors for 1-year mortality among patients with cardiac implantable electronic device infection undergoing transvenous lead extraction: the impact of the infection type and the presence of vegetation on survival. Europace. 2014;16:1490–1495. pmid:25087154
- 26. Sandoe JA, Barlow G, Chambers JB, Gammage M, Guleri A, Howard P, et al. Guidelines for the diagnosis, prevention and management of implantable cardiac electronic device infection. Report of a joint Working Party project on behalf of the British Society for Antimicrobial Chemotherapy (BSAC, host organization), British Heart Rhythm Society (BHRS), British Cardiovascular Society (BCS), British Heart Valve Society (BHVS) and British Society for Echocardiography (BSE). J Antimicrob Chemother. 2015;70:325–359. pmid:25355810
- 27. Mulpuru SK, Pretorius VG, Birgersdotter-Green UM. Device infections: management and indications for lead extraction. Circulation. 2013;128:1031–1038. pmid:23979629
- 28. Nielsen JC, Gerdes JC, Varma N. Infected cardiac-implantable electronic devices: prevention, diagnosis, and treatment. Eur Heart J. 2015;36,2484–90 pmid:25749852
- 29. Chamis AL, Peterson GE, Cabell CH, Corey GR, Sorrentino RA, Greenfield RA, et al. Staphylococcus aureus bacteremia in patients with permanent pacemakers or implantable cardioverter-defibrillators Circulation. 2001;104:1029–1033. pmid:11524397
- 30. Baddour LM, Cha YM, Wilson WR. Clinical practice. Infections of cardiovascular implantable electronic devices. N Engl J Med. 2012;367:842–849. pmid:22931318
- 31. Viola GM, Awan LL, Darouiche RO. Nonstaphylococcal infections of cardiac implantable electronic devices Circulation 2010;121:2085–2091. pmid:20439783
- 32. Klug D, Wallet F, Lacroix D, Marquié C, Kouakam C, Kacet S, et al. Local symptoms at the site of pacemaker implantation indicate latent systemic infection Heart 2004;90:882–886 pmid:15253959
- 33. Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings: Duke Endocarditis Service. Am J Med. 1994:96:200–209. pmid:8154507
- 34. Li JS1, Sexton DJ, Mick N, Nettles R, Fowler VG Jr, Ryan T, et al. Proposed modifications to the Duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis. 2000;30:633–638. pmid:10770721
- 35. Knudsen JB, Fuursted K, Petersen E, Wierup P, Mølgaard H, Poulsen SH, et al. Procalcitonin in 759 patients clinically suspected of infective endocarditis. Am J Med. 2010 Dec;123(12):1121–7 pmid:20870199
- 36. Vollmer T, Piper C, Kleesiek K, Dreier J., Lipopolysaccharide-binding protein: a new biomarker for infectious endocarditis? Clin Chem. 2009;55:295–304. pmid:18832474
- 37. Hogevik H, Olaison L, Andersson R, Alestig K. C-reactive protein is more sensitive than erythrocyte sedimentation rate for diagnosis of infective endocarditis. Infection. 1997;25:82–85. pmid:9108181
- 38. Ridker PM. High-sensitivity C-reactive protein: potential adjunct for global risk assessment in the primary prevention of cardiovascular disease. Circulation. 2001;103:1813–1818. pmid:11282915
- 39. Simon L, Gauvin F, Amre DK, Saint-Louis P, Lacroix J. Serum procalcitonin and C-reactive protein levels as markers of bacterial infection: a systematic review and meta-analysis. Clin Infect Dis 2004;39:206–17. pmid:15307030
- 40. Mueller C, Huber P, Laifer G, Mueller B, Perruchoud AP. Procalcitonin and the early diagnosis of infective endocarditis. Circulation 2004;109:1707–1710. pmid:15066945
- 41. Watkin RW, Harper LV, Vernallis AB, Lang S, Lambert PA, Ranasinghe AM, et al. Pro-inflammatory cytokines IL6, TNF-alpha, IL1beta, procalcitonin, lipopolysaccharide binding protein and C-reactive protein in infective endocarditis. J Infect 2007;55:220–225. pmid:17586049
- 42. Christ-Crain M1, Müller B. Procalcitonin in bacterial infections—hype, hope, more or less? Swiss Med Wkly. 2005;135:451–460. pmid:16208582
- 43. Al-Nawas B, Krammer I, Shah PM. Procalcitonin in diagnosis of severe infections. Eur J Med Res. 1996;1:331–333 pmid:9364034
- 44. Rawczynska-Englert I, Hryniewiecki T, Dzierzanowska D, Evaluation of serum cytokine concentrations in patients with infective endocarditis J Heart Valve Dis. 2000;9:705–709. pmid:11041188
- 45. Kern WV, Engel A, Schieffer S, Prümmer O, Kern P, Circulating tumor necrosis factor alpha (TNF), soluble TNF receptors, and interleukin-6 in human subacute bacterial endocarditis, Infect Immun 1993;61:5413–5416. pmid:8225616
- 46. Mierzchala M, Krzystek-Korpacka M, Gamian A, Durek G. Quantitative indices of dynamics in concentrations of lipopolysaccharide-binding protein (LBP) as prognostic factors in severe sepsis/septic shock patients—comparison with CRP and procalcitonin. Clin Biochem. 2011;44:357–363. pmid:21291877
- 47. Endo S, Inada K, Ceska M, Takakuwa T, Yamada Y, Nakae H, et al. Plasma interleukin 8 and polymorphonuclear leukocyte elastase concentrations in patients with septic shock. J Inflamm. 1995;45:136–142. pmid:7583359
- 48. Nof E, Epstein LM. Complications of cardiac implants: handling device infections Eur Heart J 2013;34:229–236 pmid:23091204
- 49. Zhang X, Liu D, Liu YN, Wang R, Xie LX. The accuracy of presepsin (sCD14-ST) for the diagnosis of sepsis in adults: a meta-analysis. Crit Care. 2015 Sep 11;19:323. pmid:26357898
- 50. Li Z, Peres AG, Damian AC, Madrenas J. Immunomodulation and Disease Tolerance to Staphylococcus aureus. Pathogens. 2015;4:793–815. pmid:26580658
- 51. O'Dwyer MJ, Mankan AK, White M, Lawless MW, Stordeur P, O'Connell B, Kelleher DP, McManus R, Ryan T. The human response to infection is associated with distinct patterns of interleukin 23 and interleukin 27 expression. Intensive Care Med. 2008;34:683–91. pmid:18227999
- 52. Fjell CD, Thair S, Hsu JL, Walley KR, Russell JA, Boyd J. Cytokines and signaling molecules predict clinical outcomes in sepsis. PLoS One. 2013;8:e79207. pmid:24244449
- 53.
Elecsys BRAHMS PCT Kit manual, REF 07301715200, https://pim-eservices.roche.com/eLD/(S(oftjrleuu4mza4lanlcmabwx))/mt/en/Documents/GetDocument?documentId=7b4093ea-2e48-e611-d894-00215a9b3428
- 54. Tompkins C, Cheng A, Brinker JA, Marine JE, Nazarian S, Spragg DD, et al. Significance of leukocytosis after cardiac device implantation. Am J Cardiol. 2013;111:1608–12. pmid:23523060
- 55. Sarrazin JF, Philippon F, Tessier M, Guimond J, Molin F, Champagne J, et al. Usefulness of fluorine-18 positron emission tomography/computed tomography for identification of cardiovascular implantable electronic device infections. J Am Coll Cardiol. 2012;59:1616–25. pmid:22538331
- 56. 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;36:3075–128. pmid:26320109
- 57. Lindstrom ST, Wong EK. Procalcitonin, a valuable biomarker assisting clinical decision-making in the management of community-acquired pneumonia. Intern Med J. 2014;44:390–7 pmid:24528892
- 58. Hohn A, Heising B, Hertel S, Baumgarten G, Hochreiter M, Schroeder S. Antibiotic consumption after implementation of a procalcitonin-guided antimicrobial stewardship programme in surgical patients admitted to an intensive care unit: a retrospective before-and-after analysis. Infection. 2015;43:405–12. pmid:25588968