Risk Factor Analysis and Algorithmic Approach for Managing Deep Brain Stimulation Device-Related Wound Complications

Article information

J Wound Manag Res. 2024;20(3):241-250

Publication date (electronic) : 2024 October 31

doi : https://doi.org/10.22467/jwmr.2024.03062

Hyokyung Yoo¹, Junggyo Seo¹, Sun Ha Paek², Hak Chang^,1

¹Department of Plastic and Reconstructive Surgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

²Department of Neurosurgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

Corresponding author: Hak Chang, MD, PhD Department of Plastic and Reconstructive Surgery, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea E-mail: prshchang@gmail.com

Received 2024 July 10; Revised 2024 August 21; Accepted 2024 August 24.

Abstract

Introduction

Deep brain stimulation (DBS) is a widely recognized technique that is employed to modulate brain function in patients with movement disorders. As this technique continues to evolve, complications related to implanted hardware inevitably emerge and must be addressed. The risk of wound complications related to DBS device implantation ranges from 0% to >15% [1,2]. In particular, infection and wound dehiscence, leading to hardware exposure, have been reported at rates of up to 10.6% and 2.5%, respectively [3].

Traditionally, these wound complications have been managed by removing the entire device system to prevent further spreading of infection. As explantation of the entire device can result in prolonged interruption of patient care and potential challenges for reimplantation, more recent studies have reported attempts to salvage the whole or part of the DBS system while resolving the wound issue [1,2]. Despite these efforts, many patients undergo numerous wound revisions with prolonged intravenous antibiotics, and an eventual device explanation if the trial fails. Currently, knowledge on how to best manage device-related wound complications is limited, and management of most complications rely on the surgeon’s training and personal anecdotal experience [2,4].

This study aimed to investigate the characteristics of wound complications associated with DBS device implantation in a consecutive series of patients, and provide an evidence-based algorithm for proper management by elucidating the risk factors associated with device explantation.

Methods

Management of the device-related wound complications

A single senior plastic surgeon performed all the revisions. Preoperative radiography was used to determine the path of the IPG rod, ensuring that no damage occurred during surgery (Fig. 1). All patients underwent wound debridement with total or partial capsulectomy, while some underwent device repositioning (horizontal movement or surgical plane change) or wound closure using local flaps. In cases where device explantation or repositioning was necessary, the procedure was carried out as a collaborative operation with the neurosurgeons. During each surgery, intraoperative cultures were collected for microbiological analysis, and perioperative intravenous empirical antibiotics were administered for 2–6 weeks based on consultations with infectious disease specialists. Device explantation was performed as a last resort after thorough discussion with the neurosurgeons. When necessary, the generator, extension wire, and intracranial lead on the affected side were removed.

Fig. 1.

Preoperative radiography. Preoperative radiographs were used to determine the path of the rod. (A) Skull anterior-posterior (AP). (B) Skull lateral. (C) Skull Towne’s. (D) Chest AP.

Results

DBS device implantation surgery

A total of 398 patients underwent DBS device implantation, comprising 796 electrodes and IPGs (Table 1). Electrodes were positioned for target stimulation of the bilateral subthalamic nuclei during a single-stage operation. The patient cohort consisted of 202 women (50.8%) and 196 men (49.2%), with a mean age of 59.7±6.0 years; mean body mass index was 22.8±4.0 kg/m2. There were 58 patients with diabetes (14.6%), 49 with hypertension (12.3%), and three were smokers (0.8%). During the mean follow-up of 9.8±5.5 years, 28 patients (7.0%) presented with device-related wound complications that required reoperation.

Table 1.

DBS device implantation surgery (2005–2022)

Device-related wound complications that required surgical revision

The mean age of the patients with wound complications was 64.9±6.0 years, which was slightly higher than that of the total patient cohort, although the difference was not statistically significant (Table 2). The complication group had a higher proportion of males (21:7), whereas the overall patient population was nearly evenly split between males and females (196:202) (P<0.001). No significant differences were observed in patients’ underlying diseases.

Table 2.

DBS device-related wound complications that required surgical revision (n=28)

Among a total of 28 cases of wound complications, there were 14 cases of infection (50.0%), 22 of dehiscence (78.6%), five of skin erosion (17.9%), and two of device displacement (7.1%). Most complications occurred on the scalp where the protruded cap was located (18 cases, 64.3%), followed by the neck/chest where the IPG was implanted (8 cases, 28.6%), or both the scalp and neck/chest (2 cases, 7.1%), with no cases of intracerebral involvement (0.0%). The median time from DBS insertion to reoperation for complications was 6.4±5.2 years, and the mean follow-up period was 10.3±4.7 years. The intraoperative microbiological cultures showed no growth in half of the cases (14/28 cases, 50.0%); methicillin-resistant Staphylococcus aureus (MRSA) in six cases (21.4%); methicillin-susceptible S. aureus (MSSA) in four cases (14.3%); and other skin flora, including Staphylococcus epidermidis and Propionibacterium acnes in four cases (14.3%). Various perioperative antibiotics, including cefazolin (13 patients, 46.4%), cefepime (2 patients, 7.1%), ciprofloxacin (4 patients, 14.3%), vancomycin (5 patients, 17.9%), and vancomycin+cefepime (4 patients, 14.3%), were used.

The surgical management was generally aimed at wound debridement and revision of the skin flap (28 cases, 100.0%). Various local flaps, including advancement, rotational, and transpositional flaps, were used for wound closure in 22 cases (78.6%) where an insufficient amount of skin flap remained in the original wound (Fig. 2). In 17 cases (60.7%) which the covering soft tissue provided inadequate protection for the exposed device, device replacement in different horizontal positions or into a different surgical plane was performed (Fig. 3). Although attempts were made to perform complete capsulectomy of the original device pocket whenever possible, this was achieved in only seven cases (25.0%); the other cases were managed by partial capsulectomy.

Fig. 2.

Patient with wound dehiscence in the scalp area. Wound revision was performed using a local flap and total capsulectomy. After deep brain stimulation device implantation, the patient experienced wound dehiscence at the scalp area and exposure of the implantable pulse generator rod 2 years post-surgery. The patient underwent debridement and wound revision 3 times previously. To salvage the device, device repositioning and local flap surgery using the surrounding soft tissue were performed. However, wound dehiscence recurred 4 years post-surgery. Before the 4th surgery, a mild elevation in C-reactive protein levels (0.73) was observed, and continuous methicillin-resistant Staphylococcus aureus growth was identified from surgical site cultures. The patient was administered vancomycin for 3 weeks during hospitalization. Due to persistent wound dehiscence and subsequent infection, the device was ultimately removed. (A) Preoperative image. (B) Removed capsule. (C) Postoperative image at 6 months, showing the wound remains successfully healed without dehiscence.

Fig. 3.

Patient with wound dehiscence in the anterior chest area. The implant position was changed from the premuscular to the intramuscular plane. (A) Preoperative image showing the implantable pulse generator (IPG) rod in the premuscular plane. (B) Intraoperative image showing the IPG rod buried in the intramuscular plane.

While almost half of the patients (13 patients, 46.4%) underwent only a single reoperation to resolve wound complications, nine underwent 2–3 reoperations (32.1%), and six patients underwent 4–8 reoperations (21.4%). DBS device salvage was successful in 16 patients (57.1%) and failed in 12 (43.9%), who eventually underwent device removal on the affected side. Of these 12 patients, four (14.3%) underwent reinsertion of the device at a mean of 8±4.5 months after explantation.

Risk factors associated with DBS device removal

A comparison of patient demographics between the device removal group (DR) and the device salvage group (DS) indicated no significant differences (Table 3). In comparison of wound pathophysiology, the rate of infection was significantly higher in DR (9/12 cases, 75.0%) than in DS (5/16 cases, 31.3%) (P=0.010). Meanwhile, a higher proportion of wound dehiscence (14/16 cases, 87.5%; 8/12 cases, 66.7%; P=0.085) and skin erosion (3/16 cases, 18.75%; 1/12 cases, 8.3%; P=0.198) was observed in DS than in DR. The majority of wounds were found on the scalp both in DR (9/12 cases, 75.0%) and DS (9/16 cases, 56.3%), and a large proportion of DS was also found in the neck/chest (7/16 cases, 43.8%) (P=0.236). While no significant difference in the mean follow-up period was observed between the two groups, DS showed a significantly longer time from device implantation to complications than DR (7.9±5.7 years, 4.2±3.8 years; P=0.022).

Table 3.

Risk factors associated with DBS device removal

The CRP values measured before surgical revision were significantly higher in DR (0.6±1.1) than in DS (0.1±0.1) (P=0.026). In the intraoperative culture results, the majority of DS showed no growth (11/16 cases, 68.8%), whereas in DR, 41.7% (5/12 cases) had MRSA, 16.7% (2/12 cases) had MSSA, and 16.7% (2/12 cases) had other skin flora (P=0.044). Perioperative intravenous antibiotics were administered to all patients, and there were no significant differences in the regimen or duration. The number of reoperations was significantly different between DS and DR: 81.3% of the DS (13/16 cases) underwent a single revision, whereas all patients in the DR (6/12 cases, 50.0%, 2–3 times; 6/12 cases, 50.0%, 4–8 times) underwent more than two revision surgeries (P<0.001). The rate of total capsulectomy was significantly higher in the DS (6/16 cases, 37.5%) than in the DR (1/12 cases, 8.3%) (P=0.032).

Discussion

While DBS of the subthalamic nucleus in Parkinson disease has been accepted as an effective treatment for patients with medication-associated side effects [5], complications have been reported. Although the risk of major morbidity or mortality is low, minor complications can still be stressful for patients because of the series of wound revisions, long-term use of antibiotics, long hospital stays, and eventual removal of hardware, which can result in a significant physical and emotional burden, as well as excessive resource expenditure [6]. However, previous reports on the complications of DBS surgery have not adequately addressed the patterns or appropriate management of these adverse events [1,7-10].

In this study, DBS wound complications were found in 7.0% of all the patients, mostly presenting as wound dehiscence (78.6%) accompanied by infection (50.0%) in the scalp (64.3%), neck/chest (28.6%), or both (7.1%), without involving problems in the intracerebral lesions. The overall complication rate was within the range reported in previous studies (0%–15.0%) [1-4,11]. While total hardware removal is recommended for managing intracranial infections [9,12], a more conservative treatment is recommended for wounds limited to soft tissues [10,13-15]. Since no cases involving intracerebral lesions were present in this study, the primary aim of the revision was to salvage the device and avoid unnecessary removal, which imposes a significant burden of returning to previous painful symptoms [16]. More than half of the patients with complications (53.6%) underwent multiple reoperations, and device salvage was successful in 57.1% of reoperation cases. Although the success rate was higher than the rates previously reported (20%–66%) [17-21], another 42.9% experienced numerous revision surgeries and eventual explantation of the device.

To elucidate the factors that determine the final course of revision procedures, risk factors associated with device removal were analyzed. In this analysis, the DR most commonly presented with infections (75.0%), evidenced by positive culture results and elevated CRP levels compared to the DS. Specifically, MRSA was the most common pathogen found in the DR (41.7%), and intravenous vancomycin was administered. In contrast, the DS most commonly presented with wound dehiscence (87.5%) and a relatively low rate of infection (31.3%), with the majority of cultures showing no growth and lower CRP levels. Although previous studies have reported widely varying frequencies of infectious complications, ranging from 0% to 15%, the majority are known to be associated with S. aureus [16,17,21]. In a retrospective review of 588 DBS implantation procedures, S. aureus infection was likely to exhibit early and aggressive clinical behavior, eventually leading to hardware removal [17]. A strong correlation between infection, especially MRSA, and device removal was evident in our study. Therefore, vancomycin may be considered the empirical antibiotic of choice when severe infection is clinically suspected. Furthermore, device removal should be considered if MRSA is detected during the first revision.

The wounds in DR mostly presented on the scalp (75.0%), whereas those in DS were similarly distributed along the scalp (56.3%) and neck/chest (43.8%). Particularly on the scalp, even minor wounds have a lower chance of resolution because of low soft tissue availability and high skin flap tension from previous operations. Recurrent non-healing can cause chronic skin erosion and lead to exposure to bacterial pathogens, eventually resulting in device removal. The time from device implantation to complication ranged from 0.1 to 15.6 years with a mean of 6.4±5.2 years, which was markedly longer than those reported previously (0–45 months) [7,15,17,19]. The prevalence of delayed wounds suggests that bacterial pathogens may enter through defects in the skin barrier caused by chronic and recurrent skin erosion, rather than through initial implantation [7,9,13,15]. In addition, the time from implantation to complication was significantly longer in DS than in DR. As DS had greater involvement of the neck and chest, the overlying skin flaps provided better protection and experienced less skin tension compared to DR, preventing wound complications for a longer period. Therefore, patients with DBS should avoid even minor trauma, especially to the scalp. Surgeons should implant the device at an appropriate depth and try to maintain the health of the covering skin flap.

In DS, 81.3% were successfully salvaged with only a single reoperation. In contrast, in DR, half of the patients underwent more than three reoperations before eventual device explantation. In other words, if complete wound healing is not achieved in the first few revisions, there is a high risk of device removal. Considering the time, effort, and costs required for numerous reoperations, temporary device removal at an early stage should be considered for high-risk cases. Several reports suggest that reoperations with removal of only the problematic components and preservation of the intracranial electrodes are safe, minimize costs, and improve quality of life [18,20]. However, in this study, only total removal of the device on the affected side was performed. This was necessary because the device was physically connected from the intracranial electrode to the outside extension wire and IPGs, allowing the bacteria to spread readily along the wire into the intracranium [19].

The only surgical strategy that positively affected device salvage was total capsulectomy. Complete removal of the capsule, including debris from chronic inflammatory reactions, helps resolve the infection and results in complete wound healing. Previous studies have shown that S. aureus can form surface-associated biofilms that render it highly resistant to host defense mechanisms and antibiotics, thus promoting persistent infections [22,23]. Thus, total capsulectomy is important for efficient use of antibiotics after revision.

Drawing on our experience, an algorithmic approach was designed to treat DBS-related wound complications (Fig. 4). Unless the wound involves intracerebral lesions, localized wound problems can be initially managed by simple debridement using total capsulectomy and intravenous antibiotics. If this strategy is unsuccessful, debridement can be attempted again in cases where any of the risk factors, including infection, MRSA growth, CRP elevation, a shorter time from device implantation to complication, and a history of more than three previous reoperations, are not applicable. However, in cases where any of the risk factors are applicable, temporary explantation of the problematic device should be considered at a relatively early stage, rather than performing another salvage trial. Replacement can be performed after a minimum of 3 months [24].

Fig. 4.

Wound treatment algorithm. Algorithmic approach for the treatment of deep brain stimulation (DBS) device-related wound complications. MRSA, methicillin-resistant Staphylococcus aureus; CRP, C-reactive protein.

The paper by Ginalis et al. [18] concluded that wound dehiscence in most cases can be successfully managed without complete removal of the DBS system. In contrast, our study retrospectively identifies risk factors leading to device removal and suggests performing total explantation of the device at an earlier stage when these risk factors are present. While it is preferable to salvage the device as much as possible after a wound complication occurs, this approach can lead to frequent hospitalizations and increased costs, which can be burdensome for the patient. Additionally, continuous debridement, particularly in the scalp area, can result in a shortage of skin needed for coverage, causing wound complications to recur even after more than three surgeries. Therefore, we have developed an algorithm to determine the necessity for device removal at an earlier stage to minimize the number of surgeries required for the patient. Furthermore, we believe that early total explantation of the device helps prevent the spread of bacteria to the intracranial electrode connected to the extension wire and IPG. We consider early explantation to be one of the reasons why intracerebral infection did not occur in our tertiary neurosurgery center.

This study has several limitations. First, it was a retrospective study conducted at a single institution. Additionally, the management of complications was exclusively handled by a single senior plastic surgeon. This could potentially limit the generalizability of the research findings. Second, although this case series was larger than most other studies reporting similar hardware complications with extended follow-up periods, it included a small patient sample size. Third, as the recorded wound complications were limited to cases that underwent revision in the plastic and reconstructive departments only, the exact frequency and duration of complications before seeking revision in the plastic department were unclear. Further studies with larger cohorts are warranted to establish more accurate analyses and develop better strategies to address the challenges related to DBS surgery.

In conclusion, this long-term follow-up review revealed that the DBS device-related wound complications occurred in a significant number of patients (7.0%), and that the device was salvaged successfully in 57.1% of patients. To our knowledge, this is the first study to report the risk factors that lead to eventual device removal, including infection, MRSA growth, shorter time interval from device implantation to complications, CRP elevation, and a history of multiple previous revisions. To maximize the therapeutic benefit of the device and minimize the time, effort, and cost required for surgical revision, this novel algorithmic approach, informed by our accumulated experience, suggests early temporary device removal in high-risk cases.

Article information Continued