Negative Pressure Wound Therapy With Instillation and Dwell Time Used to Treat Infected Orthopedic Implants: A 4-patient Case Series

Infection following orthopedic implants for bone fixation or joint replacement is always serious and may require removal of the osteosynthetic material. Negative pressure wound therapy with instillation and dwell time (NPWTi-d) is an emerging therapy for the treatment of complex wounds, including infected wounds with osteosynthetic material. The purpose of this case study was to evaluate the outcomes of 4 patients (1 man, 3 women; age range 49 to 71 years) with a postoperative wound infection (POWI) following fracture repair and internal fixation. All patients were at high risk for surgical complications, including infections. Standard infection treatments (antibiotics) had been unsuccessful. Based on the available literature, a NPWTi-d protocol was developed.Following surgical debridement, wounds were instilled with polyhexanide biguanide with a set dwell time of 15 minutes, followed by continuous NPWTi-d of -125 mm Hg for 4 hours. The system was changed every 3 to 4 days until sufficient granulation tissue was evident and negative pressure without instillation could be used. Systemic antibiotics were continued in all patients. Granulation tissue was found to be sufficient in 12 to 35 days in the 4 cases, no recurrence of infection was noted, and the osteosynthesis material remained in place. No adverse events were observed. Research is needed to compare the safety and effectiveness of this adjunct treatment in the management of challenging wounds to other patient and wound management approaches. 

Orthopedic implants are used mainly for bone fixation and joint replacement.^1,2 As Gustilo et al³ described in their review, infection is a serious complication; after internal fixation, the incidence of infection varies between 0.4% and 16.1% depending on the type of fracture and whether it is closed or compound.⁴ Infection of the implant is undesirable because of the intensive treatment, high financial costs, and the uncertain patient outcome. A literature review⁵ has shown the infection can result in (functional) loss of the affected limb. 

An extensive literature review⁶ has shown implanted hardware is highly likely to cause bacterial infections, a response associated with the host defense to the foreign body. The most frequently isolated bacteria causing these implant-related infections are Staphylococcus aureus and coagulase-negative S epidermidis.⁷ When an orthopedic implant-related infection occurs, the first choice of treatment is to remove the implant.⁴ However, when a fracture has not yet healed, most surgeons prefer to leave the implant in place to allow the fracture to heal and perform one or more surgical debridements with or without topical antibiotic treatment.⁴ 

Reviews of the literature^5,6 have shown a proteoglycan-containing biofilm covers the implant and acts as a carrier for the pathogens, causing the infection.^5,6 This makes complete eradication of the pathogen a challenge, because many antibiotics administered intravenously cannot penetrate this biofilm.⁶ When used for bone fixation, the implants are needed only temporarily and can be removed when the fracture is healed. Removing the implant will eradicate the underlying cause of an ongoing infection.⁴

In 1997, Morykwas et al⁸ published a prospective cohort study (N = 10) describing a new wound management method that involved the application of subatmospheric pressure to human wounds (ie, negative pressure wound therapy [NPWT]). Although the review by Morris et al⁹ shows significant evidence in general for this treatment modality is lacking, NPWT has gained wide acceptance in the treatment of acute and chronic wounds. 

VAC-Instill therapy (KCI, an Acelity Company, San Antonio, TX) is a combination of traditional NPWT and intermittent instillation (NPWTi-d) of a topical wound treatment solution (ie,  NPWTi-d allows a combination of NPWT with timed, intermittent delivery of topical solutions). The NPWTi-d unit has a patch with 2 lumina: 1 to maintain the negative pressure and 1 for delivering the fluid into the foam that is in contact with the wound bed. The NPWTi-d cycles through 3 phases: the vacuum phase, the instill phase (dwell time), and in-between phase. The vacuum phase, which can range from 1 to 12 hours, establishes the vacuum. During the instill phase, the fluid is allowed to flow through the foam into the wound. The dwell time differs depending on the type of fluid. In between the instill phase and vacuum phase, the fluid is suctioned out from the foam and wound; this phase lasts 30 seconds and immediately precedes the vacuum phase.

As shown in a clinical observational study (N = 32) by Lehner et al,¹⁰ adding instillation to NPWT may offer additional benefits compared to NPWT alone in reducing the bacterial biofilm. Furthermore, a controlled trial¹¹ (N = 74) showed the therapy increases the granulation tissue (P <0.05) compared to NPWT without instillation, and irrigates the wound in a sealed environment, potentially preventing cross-contamination events (P <0.05).   

In 1998, Moch et al¹² were among the first to describe use of NPWTi-d in the treatment of wound infections and osteomyelitis. In this study, 27 patients diagnosed with osteomyelitis were provided instillation with vacuum sealing. None showed an infection after 3 to 14 months. In a 2009 retrospective, case-control cohort study, Timmers et al¹³ reported successful use of NPWTi-d to treat post-traumatic osteomyelitis (N = 30). The NPWTi-d group had a significantly lower recurrence of infection (P <0.0001) of the treated body part compared to the group treated using a wound dressing with gauze. NPWTi-d treatment of infected orthopedic joints, first described by Lehner et al,¹⁰ showed that in 26 (more than 80%) of the patients, removal of the prosthesis was not necessary at 4 to 6 months’ follow-up after treatment with NPWTi-d. 

Given this research, the authors developed a treatment protocol using NPWTi-d for patients with infected orthopedic implants. The purpose of this case study was to evaluate the outcomes of this protocol of care. 

See Table 1.

Case 1. Mr. H, a 71-year-old patient, presented to the emergency department with a history of hypertension controlled with lisinopril. He had fallen from a tall tree 7 days before presentation. An x-ray and CT scan showed an uncomplicated supracondylar humeral fracture with dislocation (AO-classification 13- C1). An open reposition and internal fixation was performed 17 days after the accident, delayed due to severe soft tissue swelling. Fixation was performed using double plating with distal humeral plates (DePuy Synthes, Zeist, The Netherlands). 

Six (6) days after surgery, Mr. H presented to the emergency department with a fever and pain at the surgical site. Infection parameters were elevated (CRP 211 mg/L, leukocytes 10.8 10^9/l). Due to the combination of pain, illness, and elevated infection parameters in laboratory tests, a postoperative wound infection (POWI) was diagnosed. Mr. H was admitted to the hospital and antibiotic therapy using flucloxacillin IV, 6 g/24 hours, was started. Despite 3 days of antibiotic therapy, surgical exploration was required around the fixation site due to ongoing redness, swelling, and pain. Pus was drained, and the wound edges were approximated over gentamycin beads. A wound culture showed S aureus sensitive for flucloxacillin. 

Three (3) days after removal of pus at the surgical site, the gentamycin-beads were removed and NPWTi-d was applied. The wounds were instilled as described. After 5 days of NPWTi-d, rapid granulation of the wound was seen (see Figure 1). At that time, the infection parameters had returned to almost normal values (CRP 31 mg/L, leukocytes 6.1x10^9/L). After 12 days, the NPWTi-d was changed to a NPWT system and Mr. H was discharged 21 days after re-admission (see Figure 2). The NPWT was continued at home without antibiotic therapy. After 1 month, a split skin graft was performed to close the wound. The skin graft was managed using NPWT for 5 days. Epithelialization of the wound and good adhesion of the graft (see Figure 3) were noted after removal of the NPWT system. Follow-up at 6 months showed no further complications, with a good function of the elbow. No adverse events occurred during wound treatment. 

Case 2. Otherwise healthy 49-year-old Ms. K fell off her bike onto the sidewalk. (A few years before this accident, she had a correction of her hammertoes.) She suffered an uncomplicated fracture of the tibia plateau (Schatzker classification 6). Nine (9) days after the trauma, she underwent open reposition and internal fixation using a tibia-locking compression plate. The postoperative x-ray and CT scan 1 day after surgery showed a persistent depression of the lateral tibia plateau, necessitating reoperation during which the depression was corrected and the plate replaced. The skin was closed over a gentamycin-collagen resorbable dressing using staples. The surgeon provided 3 doses of antibiotics postoperatively (cefazolin IV, 1500 mg, over 24 hours) because of the estimated higher risk of infection; this was not a standard procedure.  

When Ms. K went for follow-up in the outpatient clinic 14 days later, a wound infection was diagnosed. Wound cultures showed S aureus infection with no resistance for flucloxacillin. Multiple surgical wound debridements, NPWT, and long-term antibiotic treatment over 4 months had no effect. Despite lateral gastrocnemius transposition to close the wound, the infection persisted. 

Six (6) months later, Ms. K was admitted with a persistent infection of the tibial plate (see Figure 4). Surgical debridement was performed and NPWTi-d was used to cover the wound combined with antibiotic therapy using intravenous flucloxacillin, 6 g/day, for 5 weeks. The wound was instilled with polyhexanide biguanide, and NPWTi-d was provided as described. After 3 weeks, the wound was fully granulated (see Figure 5) and NPWT continued without the instill option. After 2 more weeks, the wound was closed (see Figure 6). Infection did not recur during the follow-up, and 6 months later the orthopedic surgeon successfully placed a total knee prosthesis, due to functional limitations. Six (6) months of follow-up by the orthopedic surgeon showed no infection or complications of the knee prosthesis placement. No adverse events occurred during wound treatment. 

Case 3. Ms. L, a 64-year-old with a medical history including alcohol abuse and Korsakoff syndrome with severe memory impairment, presented to the emergency room 6 weeks after an inversion trauma of the left ankle. Her only noted medication was thiamine; her health was otherwise normal. 

X-rays showed a trimalleolar ankle fracture. Ms. L was admitted to the hospital, and conservative treatment with cast therapy was started due to severe soft tissue swelling. During admission, Ms. L’s condition improved and an open reduction and internal fixation was performed using a tritubular plate (DePuy Synthes, Zeist, The Netherlands) on the lateral side. No fixation on the medial side was performed due to soft tissue problems. A cast was placed postoperatively, and Ms. L was advised to avoid weight-bearing while ambulating. Despite this advice, Ms. L stood on her left ankle and x-ray studies showed a progressive subluxation of the talar bone with bowing of the fibular plate. 

Eighteen (18) days after the first operation, redo surgery was performed using 2 plates on top of each other on the fibula and an extra Drittelrohr plate at the dorsal side of the fibula. The medial fracture was fixated with tension band wiring. The lateral wound could not be closed due to soft tissue problems. NPWTi-d was placed (see Figure 7) and the wound was instilled with polyhexanide biguanide, with dwell time and NPWT provided as described. 

Wound cultures of the medial wound showed normal dermal bacteria (ie, no specific growth of any type). Intravenous prophylactic antibiotic treatment was started using flucloxacillin, 4 g/day. After 4 weeks, granulation tissue completely covered the plate. Instill therapy was changed to NPWT (see Figure 8). Two (2) weeks later, NPWT was discontinued because the wound was fully granulated.

Ms. L died a few weeks after she was discharged from the hospital. It was believed chronic alcohol abuse and cerebral complications of Korsakoff syndrome led to her death. In this patient, Korsakoff syndrome was a big risk factor in developing a wound infection. Because she could not remember to avoid weight-bearing on her operated ankle, the implant was bent, necessitating a second operation, which is associated with a higher risk of infection.¹⁴ No adverse events occurred during wound treatment. 

Case 4. Ms. M is 59 years old and presented to the outpatient orthopedic clinic with a pressure ulcer at the medial malleolus, a compression ulcer related to the valgus position of the foot. Ms. M had undergone an ankle prosthesis placement due to rheumatoid arthritis (for which she used methotrexate) 5 years before. Additional medication included aspirin, a beta-blocker, and a statin. Ms. M also suffered from a normocytic anemia caused by chronic illness. On presentation, she used carbasalate calcium, a beta-blocker, a statin, omeprazole, and vitamin B tablets. 

Corrective osteotomy was performed through medial and lateral incisions. Prophylactic antibiotics were provided postoperatively for 24 hours. After 4 weeks, a wound infection developed at the lateral surgical wound. The plate was visible after surgical debridement (see Figure 9). Antibiotic treatment was started using intravenous flucloxacillin, 6 g/day. Wound cultures showed S aureus and an Escherichia coli infection. 

NPWTi-d therapy was started on the lateral wound. The hypothesis was the NPWTi-d would be able to temper bacterial counts and reduce the biofilm presumed to be present on the hardware. If the infection could not be treated, amputation of the lower leg was the last resort. The wound was instilled with polyhexanide biguanide, and dwell time and NPWT were provided as previously described during admission. Due to an allergic reaction to the flucloxacillin, Ms. M’s antibiotic therapy was switched to vancomycin. The medial and lateral wounds closed without complications and demonstrated granulation tissue (see Figures 10 and 11). After 5 weeks of NPWTi-d, therapy was changed to a low pressure, portable NPWT system for another 120 days until full closure (see Figure 12). The antibiotic treatment was continued for 6 months. After 10 months of follow-up, both wounds were fully closed without signs of infection (see Figure 13). No adverse events occurred during wound treatment. 

Wound infections related to orthopedic implants are difficult to treat, a burden for the patient,¹⁵ and associated with high health care costs.^16,17 When an infection cannot to be contained, implants have to be removed. If the implant is used for bone fixation, the fracture may not be fully healed. Therefore, it is important to develop therapeutic strategies for treating these infections. This study describes 4 cases in which systemic antibiotics, mechanical wound debridement, and NPWTi-d were used to manage wounds. In all 4 cases, the infection resolved without implant removal. On average, patients received NPWTi-d treatment over a period of 12 to 35 days. In the authors’ opinion, NPWTi-d must be used in combination with mechanical debridement and systemic antibiotic therapy when used to manage infected wounds with implants and it should be seen as an adjunct to standard therapy. This belief is supported by evidence from the literature, albeit most of it consisting of small case series and reviews. However, these first experiences showed promising results. 

 The wounds were instilled to fill the foam with a set dwell time ranging from 12 to 15 minutes, followed by continuous NPWT of -125 mm Hg for 4 to 8 hours. The dwell time of 13 to 15 minutes corresponds with a 2013 consensus document.¹⁸ A barrier spray was used to protect the surrounding skin. In additional cases with other wounds, the authors noticed maceration on the surrounding skin after using the instillation with a frequency above 6 times per 24 hours and subsequently initiated NPWTi-d at 4 to 6 times per day. As previously mentioned, no literature was found to guide the frequency of instillation and level of negative pressure. The minimum period of treatment was 2 weeks. The NPWTi-d system was changed 2 times a week, every 3 or 4 days. No leakage of the system during therapy required the authors to change the system more often. No therapy-related complications occurred. 

The authors hypothesized the main advantage to using NPWTi-d is that the wound is flushed with an antiseptic solution, in this study polyhexanide biguanide, which is a more effective wound debridement protocol than NPWT alone. Different antiseptic solutions for wound irrigation (eg, polyhexamine bigluconate [PHMB], saline, or antibiotics) are a consideration. Relevant evidence regarding dwell time, the level of negative pressure, and the frequency of the instillation is not described in the literature. The authors chose polyhexanide biguanide (PHMB + Betaine solution) as the wound irrigation solution. An in vitro study by Ikeda et al¹⁹ showed an advantage to using PHMB for the treatment of soft-tissue injuries without bone involvement compared to povidone-iodine and silver nitrate. The use of a PHMB solution is advised^16,20 when treating infections similar to those described in this case study. A small in vitro study²⁰ showed wound dressings combined with polyhexanide biguanide lowered bacterial activity in 32% of the tested samples, but some colonies of S aureus were not eradicated when polyhexanide biguanide was added. The authors believe the physics of flushing in and of itself may facilitate more rapid wound closure; the contribution of specific agents such as PHMB may be only marginal. 

In these study cases, 3 out of 4 patients suffered from an infection caused by S aureus and 1 exhibited E coli. The effectiveness of PHMB against these types of gram-negative bacteria is not well described in the literature. More research is needed to identify the best antiseptic fluid when using NPWTi-d and compare effectiveness of this method to other irrigation techniques. 

This study involves only 4 cases. Although the authors use NPWTi-d in a variety of other wounds, the number of patients with infected osteosynthesis material that need elaborate therapy is low in their clinic. In addition, the supportive information provided by the literature is low-level evidence, such as small case reports and case series of in vitro studies; large, randomized, controlled trials have not yet been performed.  

Dr. Dettmers is a surgical resident in Trauma Surgery, Alrijne Hospital Leiderdorp; and a surgical resident in wound care, Alrijne Woundcentre Leiderdorp, The Netherlands. Dr. Brekelmans is a wound specialist; Dr. Leijnen is a Trauma Surgeon; and Dr. van der Burg is a vascular surgeon, Alrijne Woundcentre Leiderdorp. Dr. Ritchie is a Trauma Surgeon, Alrijne Hospital Leiderdorp.

Please address correspondence to: R.C. Dettmers, Alrijne Ziekenhuis Leiderdorp Dept General Surgery, Simon Smitweg 1, 2353GA Leiderdorp, The Netherlands; email: R.C.dettmers@lumc.nl.