Do Antibiotic-Impregnated Envelopes Prevent Deep Brain Stimulation Implantable Pulse Generator Infections? A Prospective Cohort Study

Abstract

Introduction

Infection after deep brain stimulation (DBS) implanted pulse generator (IPG) replacement is uncommon but when it occurs can cause significant clinical morbidity, often resulting in partial or complete DBS system removal. An antibiotic absorbable envelope developed for cardiac implantable electronic devices (IEDs), which releases minocycline and rifampicin for a minimum of 7 days, was shown in the WRAP-IT study to reduce cardiac IED infections for high-risk cardiac patients. We aimed to assess whether placing an IPG in the same antibiotic envelope at the time of IPG replacement reduced the IPG infection rate.

Methods

Following institutional ethics approval (UnitingCare HREC), patients scheduled for IPG change due to impending battery depletion were prospectively randomised to receive IPG replacement with or without an antibiotic envelope. Patients with a past history of DBS system infection were excluded. Patients underwent surgery with standard aseptic neurosurgical technique [J Neurol Sci. 2017;383:135–41]. Subsequent infection requiring antibiotic therapy and/or IPG removal or revision was recorded.

Results

A total of 427 consecutive patients were randomised from 2018 to 2021 and followed for a minimum of 12 months. No patients were lost to follow-up. At the time of IPG replacement, 200 patients received antibiotic envelope (54 female, 146 male, mean age 72 years), and 227 did not (43 female, 184 male, mean age 71 years). The two groups were homogenous for risk factors of infection. The IPG replacement infection rate was 2.1% (9/427). There were six infections, which required antibiotic therapy and/or IPG removal, in the antibiotic envelope group (6/200) and three in the non-envelope group (3/227) (p = 0.66).

Conclusion

This prospective randomised study did not find that an antibiotic envelope reduced the IPG infection rate in our 427 patients undergoing routine DBS IPG replacement. Further research to reduce IPG revisions and infections in a cost-effective manner is required.

Introduction

IPG infection is a serious complication of DBS surgery, with an incidence in the order of 4.9% [1] and resultant considerable patient morbidity and health care cost [2]. New strategies to reduce IPG revision and infection are required. One potential such strategy is the use of an antibiotic-impregnated TYRX™ envelope. Use of this envelope reduced infection in high-risk cardiac patients with IED in a large trial (WRAP-IT) [3]. We aimed to assess whether this envelope reduced infection rates for patients undergoing replacement of depleted IPGs. We report a cohort of patients who were prospectively randomised to receive either an antibiotic envelope or no envelope at the time of IPG replacement.

Methods

All patients were operated on at either the Wesley Hospital or St Andrew’s War Memorial Hospital, Brisbane, Australia, by or under supervision of the senior surgeon (TC). Inclusion criteria were all adult DBS patients scheduled for IPG change due to impending depletion. Exclusion criteria were patients with a past history of DBS system infection or allergy to the envelope antibiotics minocycline or rifampicin. No patients declined to participate. Patient age, the indication for DBS, previous IPG replacement surgeries, and medical co-morbidities including diabetes and hypertension were recorded. Anticoagulant and antiplatelet agents were ceased peri-operatively.

The IPG was replaced with standard neurosurgical technique in the infraclavicular suprapectoral position. All patients received 24 h peri-operative antibiotic prophylaxis (cephazolin, unless allergic to cephalosporins in which case vancomycin was used).

On the day of surgery, patients were randomised by an unblinded coin toss to receive IPG replacement with or without a TYRX™ (Medtronic) envelope. If randomised to envelope use, the IPG was placed in a large TYRX™ envelope, a polypropylene mesh that elutes the antimicrobial agents minocycline 7.6 mg and rifampicin 11.9 mg. The envelope is fully absorbed in 9 weeks. As part of routine post-surgical care, patients were monitored for signs of IPG site complications including infection or haematoma. Staples were removed on post-operative day 8.

Patients were regularly followed for a minimum of 12 months and infections requiring antibiotic therapy and/or IPG removal were recorded. Swabs were cultured and assessed for antibiotic sensitivity to assist with refining antibiotic treatment. The development of a pocket haematoma in the peri-operative period was recorded. Statistical analyses were performed with GraphPad Prism. Univariate analysis was performed with a two-sided Student’s t test and variables with p < 0.2 were selected for multivariate regression analysis with significance at p < 0.05.

Results

A total of 427 consecutive patients were randomised into the trial over the period July 2018–June 2021 and followed until June 2022. Overall, 200 patients received a TYRX™ envelope and 227 patients did not. There were no statistical differences in the two groups: age (72 years envelope vs. 71 years non-envelope), sex (54 females and 146 males envelope vs. 43 females and 184 males non-envelope), medical co-morbidities, number of previous revisions, or primary DBS indication.

Nine patients developed an infection requiring antibiotics and/or IPG removal, an infection rate of 2.1% (9/427). In the envelope group, there were 6/200 infections, which required antibiotic therapy and/or IPG removal, and in the non-envelope group there were 3/227 infections. Of patients with infection, co-variate analysis did not identify differences in the patient characteristics of age, sex, DBS indication, or number of precious DBS IPG placements.

Cellulitis, erythema, swelling, and purulent discharge were the most common presentations of infection. There were no cases of wound dehiscence.

There were two post-operative haematomas. One infection had moderate haemoserous fluid at revision and was included as a haematoma. The other haematoma was not managed as an infection and resolved with supportive therapy. No adverse clinical events were attributable to the use of the antibiotic envelope in the 200 patients who received the envelope.

Discussion

IPG infection is a common complication of DBS surgery, with an incidence of 4.9% in a recent meta-analysis of 11,289 patients [4]. The literature overall indicates an infection rate ranging from 1.2% to 15.2% [5]. The infection rate reported in this study is within this range.

Peri-operative antibiotic use is strongly associated with lower infection rate [6] and may have contributed to the relatively low infection rate in this study. Given the low infection rate, our study did not confirm potential risk factors reported in retrospective series including primary [7] or repeat implantation [8], operation time, body mass index, age, sex, diabetes [9], smoking, indication, capsulotomy, post-operative ICU admission [10], anti-thrombotic medication use [11], antibiotic wash, chlorhexidine skin preparation [12] (20), topical vancomycin powder [13], glycerol [14], or preoperative MRSA screening and treatment [15].

Pocket haematoma is reported to elevate the risk of infection [16], and use of the envelope is reported to reduce this risk in patients with haematoma [17]. However, there was a low haematoma rate in our series (2/427), similar to the rate of 0.4% in the WRAP-IT trial. A recent publication indicated a 1% haematoma rate for patients not on anti-thrombotic medication and a 5.9% haematoma rate for patients taking anti-thrombotic medication [11]. With increasing use of anti-thrombotic medication, there may be a role for assessing envelope use in this higher risk population.

Of the 9 patients who developed infection, the results of cultures were similar in the envelope and non-envelope groups. Skin commensal organisms were the most common, reflecting the literature (14, 28), including in the envelope group, despite the envelope containing minocycline and rifampicin in effective concentrations for common IPG infective organisms (30, 31).

In the group who developed infections, we explanted the IPG and extension leads just distal to their connections with the brain leads in the retroauricular scalp, and after a week of intravenous antibiotic therapy followed by 4 weeks of oral antibiotics the extension leads and IPGs were re-implanted with no subsequent re-infections or hardware complications. Our study did not find an association between IPG infection and envelope use, in keeping with a previous study, which reported that the association between envelope use and infection was not significant after controlling for confounding variables including peri-operative antibiotic administration [6]. Other retrospective studies report mixed results, with one study reporting reduced infection with antibiotic pouch use in a potentially higher risk population [18] and another study looking at primary implantation procedures that reported no IPG site infections, with the only reported infections occurring at the scalp [19]. Our no adverse event profile related to the pouch was similar to these studies. We could not confirm whether the envelope may potentially reduce twiddler’s syndrome [20] or other complications [8] such as gelatinous mass formation [21] or granulation tissue [22].

Establishing the potential efficacy of an intervention to reduce infection is complicated by low baseline rates of infection, small effect size, and resultant need for large trials [13]. This study assessed patients encountered in routine functional neurosurgical practice undergoing replacement of depleted IPG was designed prior to the results of WRAP-IT, which identified benefit of antibiotic pouch use in higher risk cardiac patients. The population in this study was not particularly high risk, and excluded those with prior infection, which is a known risk factor [18]. The infection rate was low. There are conflicting reports of contributing risk factors for infection, which were minimised by the process of randomisation. There was no blinding however, and the authors had direct responsibility for patient care. Consistent experienced surgical technique was applied across the study. Although this study included 427 patients, a larger multi-centre study in a higher risk population may address these limitations. Finally, avoiding IPG replacement may reduce associated morbidity, and alternative approaches may involve refinement of rechargeable technologies.

Conclusion

This prospective randomised study did not find that an antibiotic-impregnated envelope reduced IPG infections in 427 patients undergoing routine DBS IPG replacement, similar to the findings of a large retrospective North-American series [6]. Our study is generalisable to routine neurosurgery practice. As a result of this trial, the authors have not implemented routine use of the TYRX™ envelope in patients undergoing routine IPG replacement. Further research to reduce IPG revisions is warranted. A larger trial selective for patients at higher risk of infection may refine whether the envelope confers benefit in this challenging sub-cohort in a cost-effective manner.

Statement of Ethics

The study was approved by our institution’s ethics board (UnitingCare HREC approval number: 1805, Brisbane, Australia) and was performed in accordance with the Declaration of Helsinki 1975. All patients had written informed consent obtained.

Conflict of Interest Statement

The authors declare no conflicts.

Funding Sources

No external funding was sourced.

Author Contributions

Authors M.C., T.H., P.S., and T.C. contributed equally and fulfil the ICMJE criteria for authorship.

Funding Statement

No external funding was sourced.

Data Availability Statement

The primary research data generated or analysed that support the findings of this study have not been publicly uploaded on ethical grounds due to their potentially identifiable nature; further enquiries and reasonable requests may be considered by the research team and can be directed to the senior or corresponding authors.