Introduction

Sacral nerve stimulation is used to treat refractory cases of urge urinary incontinence, urinary frequency, and non-obstructive urinary retention. Sacral nerve stimulation may also help patients with cystitis, fecal incontinence, and chronic pelvic pain syndrome. With its expanding use, research evaluating postoperative complications is increasingly necessary.

The Interstim Sacral Nerve Stimulation System (Medtronic, Minneapolis, MN, USA) was approved by the Food and Drug Administration (FDA) in 1997. The operative approach at that time required a pretest and, if successful, an incision over the sacrum with direct fixation of a tined lead at the level of the S3 foramen. This method was associated with a 6.1% infection rate, and 33% of the patients required removal or revision of the stimulator [1, 2]. The technique was modified in 2002 and is now performed with placement of a tined lead during the pretest. If the patient has a favorable test, the tined lead is left in place and connected to an implanted neurostimulator. Review of the current literature reveals few studies that address infection or complication rates associated with this less invasive technique. The following is a retrospective chart review reporting rates of infection in 37 patients after neurostimulator placement.

Materials and methods

Between October 2002 and June 2005, 37 women underwent Interstim placement via 2 sequential procedures. To complete the first stage, patients were brought to the operating room and prepped with betadine scrub in the prone position. Monitored anesthesia care (MAC) was augmented by local injection of approximately 10 cc 1% lidocaine to provide intraoperative pain relief. A horizontal line was marked at the inferior margin of the sacroiliac joint. A second vertical line was marked through the sacral foramina. Patients were provided intravenous antibiotics before skin puncture; however, the type and dose were not standardized. A tined lead was placed through the third sacral foramen, motor response confirmed, and the lead tunneled to the ipsilateral gluteal region as per standard Interstim technique (Fig. 1). The lead was then connected to a temporary cable and subcutaneously tunneled to an exit point over the contralateral buttock. The connection sites were irrigated with antibacterial solution before closure with vicryl interrupted sutures.

Fig. 1
figure 1

A 3.5 in. needle is placed thru the third sacral foramen under fluoroscopic guidance thru which the tined neurostimulator lead is passed. Reprinted with permission of Medtronic, Inc. copyright 2006

Full size image

After stage I, patients were evaluated in the office between postoperative days 5 and 7 for response to the intervention and healing. If the patient exhibited a positive response, they were returned to the operating room within 2 weeks for the 2nd stage of the procedure. All patients demonstrated >50% reduction in their primary symptoms after stage I tined lead placement and were candidates for an implanted neurostimulator.

During stage II neurostimulator placement, patients returned to the operating room. Positioning, betadine skin preparation, and anesthesia were similar to that achieved during the first stage. Patients were provided intravenous antibiotics; however, the type and dose were not standardized. An incision was made over the previous operative site with a scalpel. A 2-cm deep pocket was created for the neurostimulator. The pocket was irrigated with antibacterial solution. The neurostimulator was attached to the tined lead and placed in the pocket (Fig. 2). The subcutaneous tissue was closed with 2-0 vicryl in a running fashion and the skin was closed with staples. Oral antibiotics were provided for 1-week post discharge and patients were evaluated at 2 and 6 weeks for wound healing.

Fig. 2
figure 2

Neurostimulator attached to tined lead completing stage II procedure. Reprinted with permission of Medtronic, Inc. copyright 2006

Full size image

Intraoperative and postoperative prophylactic antibiotic regimens were not standardized among women who underwent stimulator placement. The women received either a first-generation or second-generation cephalosporin plus antibacterial irrigation intraoperatively and either levaquin or ciprofloxacin to complete a 7-day course in the postoperative period.

Patients with evidence of infection including erythema, induration, pain or drainage were placed on oral antibiotics and instructed in wound care. If present, drainage was sent for culture. Antibiotics and wound care were attempted in each of the five cases of wound infection; however, the patient’s symptoms failed to improve. All five patients were returned to the operating room for device and lead removal. Intravenous antibiotics were provided at the time of surgery and the wounds were irrigated with antibacterial solution. Wounds were either closed or packed for closure by secondary intention. Patients were discharged home on oral antibiotics. If a patient desired reimplantation, a combined two-stage procedure was employed upon resolution of the infection.

The study was approved by the Institutional Review Board before study commencement. Statistical analyses were performed using MedCalc for Windows, version 8.2.1.0 (MedCalc Software, Mariakerke, Belgium).

Results

After stage II neurostimulator placement, 5 out of 37 (13.5%) women required device removal for culture positive wound infections. There was no statistically significant difference between those who developed infection and those who received a device but did not develop infection with regard to age or duration of stage I. Of the variables evaluated, higher BMI was the only statistically significant finding associated with higher rates of infection (p = 0.0097) (Table 1).

Table 1 Comparison of patients who developed wound infection after sacral nerve stimulator placement with patients who did not develop infection
Full size table

Point prevalences were calculated to evaluate additional host-related risk factors contributing to our study group’s intrinsic susceptibility to infection. Conditions such as diabetes mellitus, concomitant infections at the time of surgery, malignancy with immunosuppressive therapy, tobacco and alcohol use were abstracted. No individuals in the study group suffered from concurrent infections on the day of surgery nor were any of the women being actively treated with immunosuppressive therapy. The prevalence of alcohol and tobacco use was higher in the patients who developed infection (Table 2).

Table 2 Prevalence of tobacco and alcohol use
Full size table

Operative characteristics including estimated blood loss and duration of surgery were reviewed. All of the cases experienced minimal blood loss quantified as less than 5 ml. Mean blood loss in the non-infected group was 4.6 ml with a median of 4 ml. Surgical duration was also similar between groups with an average duration of 32 min in those patients who later developed infection and 34 min in the non-infected group.

Patients returned an average of 147.4 days after device implantation for device removal. Infection occurred a minimum of 33 days, a median of 76 days, and a maximum of 461 days after sacral nerve stimulator implantation. Symptoms included tenderness (80%), erythema (80%), discharge (60%), burning (20%), and swelling (20%). No patients were febrile or manifested signs of systemic bacteremia. Although a variety of conservative modalities including oral antibiotics and wound care were employed, all patients failed to resolve their infection and required surgical removal of the implant. Intraoperative wound cultures most commonly revealed Staphylococcus aureus (Table 3). After Interstim removal, all patients resolved their infections. Two patients underwent uncomplicated reimplantation in the contralateral buttock 14 and 16 days after device removal.

Table 3 Bacterial pathogens cultured from infected sacral nerve stimulator wounds
Full size table

Discussion

Sacral nerve stimulation has become an increasingly important therapy for the management of urge urinary incontinence, urinary frequency, and non-obstructive urinary retention and demonstrates promising benefit for the management of cystitis, fecal incontinence, and chronic pelvic pain. As a consequence, more than 8,000 neurostimulators have been implanted worldwide [3, 4]. Unfortunately, little data exist addressing complications, specifically implant-related infections.

In a study conducted by Pannek et al. [4], they found that 45.5% of percutaneous leads were colonized with bacteria and speculated an elevated risk of infection compared with the original technique where the lead was placed after the completion of a successful pretest. Their study concluded that the most common bacteria identified were either intestinal or skin surface flora similar to those isolated in this review. Furthermore, Pannek et al. suggested that the bacteria were likely present in the sacral canal during the initial tined lead placement. Given the stringent aseptic technique observed in this study and the perioperative antibiotic prophylaxis and duration to infection, one can similarly assume that it is unlikely that contamination occurred during the operative experience but rather reflects seeding of the device at a later stage.

Once infection was identified, all patients in this series eventually required explantation. Hijaz et al. [2] experienced a 10.7% infection rate after neurostimulator implantation. In all cases, the neurostimulators eventually required removal to manage the infection. Attempts at more conservative management and device relocation were unsuccessful.

It is unclear what predisposed some women to infection. The Centers for Disease Control (CDC) identified the most critical host-related risk factors for the development of any surgical site infection as age, obesity, disease severity, an American Society of Anesthesiologists (ASA) score greater than two, prolonged preoperative hospital stay, and infection at distal sites. Other possible risk factors include poor nutrition, diabetes mellitus, malignancy, immunosuppression, tobacco use, and alcohol abuse [5].

An obvious discrepancy between the women who suffered from infection and those who did not is the BMI (41.4 vs 30.8). Obesity may have led to poor hygiene, poor nutrition, wound breakdown, and poor healing among other factors. The infected group was also generally younger but otherwise the groups were well-matched with regard to parity and duration between stage I and stage II. Tobacco and alcohol use also appears to contribute to infection development.

The results of this study suggest that the risk of infection after tined lead pretest and neurostimulator placement may be higher than previously observed in older techniques. Infection is best managed with immediate explantation rather than antibiotic or surgical manipulation. Lastly, obesity, tobacco and alcohol use appear to be associated with an increased rate of infection. Further prospective randomized studies will be helpful to assess factors contributing to neurostimulator infection.