Case Studies

       The authors utilized a solution of 2 liters of normal saline with 500,000 units polymyxin B and 50,000 units of bacitracin in the initial 5 cases detailed below. Programming for all cases included 6 hours of NPWT at 125 mmHg followed by instillation of solution for 90 seconds and a dwell time of 5 minutes. One should note that these agents must be left in the wound for at least 60 seconds before removal to be bactericidal.
       Case study 1: diabetic foot ulcer with osteitis of the fifth metatarsal head. A 56-year-old White male with type II diabetes mellitus presented with a nonhealing diabetic foot ulcer of 4 months duration under the fifth metatarsal head of his left foot with new onset of drainage, redness, and pain. Examination revealed a deep ulcer with exposure of bone. Noninvasive vascular testing revealed adequate flow for healing.
       In the operating room, the wound was surgically debrided, including resection of the metatarsal head through an accessory incision on the day of admission. The patient was started on empiric intravenous antibiotics. Intraoperative cultures and Gram stain were negative likely due to perioperative intravenous antibiotics. Bone specimen sent to pathology revealed focus of chronic osteomyelitis and foci of periosteal new bone formation.
       Negative pressure wound therapy with instillation was utilized in conjunction with intravenous antibiotics from the fourth through sixth inpatient day. Profuse granulation in the plantar wound was noted with the collapse of the inner soft tissues and effective coverage of the metatarsal stump. The wound appeared clean and viable. The patient was discharged to the home with daily packing of the wound, oral antibiotics, and an offloading boot. After 13 weeks, the plantar wound was completely closed with no subsequent infection.
       This wound had been open for 4 months before it was surgically debrided and treated with NPWT with instillation. Following this therapy, the wound was completely closed in 13 weeks. In the authors’ opinion, NPWT with instillation stabilized this wound and directly led to faster discharge to the outpatient setting. By allowing rapid growth of soft tissue in the depths of the wound, the bone was covered. Had the bone remained exposed, inpatient treatment would likely have continued for a longer period of time.
       Case study 2: diabetic foot ulcer with osteomyelitis. A 62-year-old White male with type II diabetes mellitus presented with a left foot ulcer of 3 months duration with drainage between the fourth and fifth toes. X-rays confirmed osteomyelitis of the fourth and fifth proximal phalanges. A Doppler examination revealed adequate arterial flow. Following cardiac risk assessment, the patient underwent wide surgical debridement of the infected web space and excision of the fifth ray and osteomyelitic portions of the proximal phalange in the fourth digit on inpatient Day 4. The fourth metatarsal head was completely exposed after excision of devitalized capsule, revealing healthy cartilage and bone. Operative Gram stain revealed no organisms and culture of excised bone revealed Corynebacterium striatum. Pathology revealed fragments of bone with focal and acute osteomyelitis. Negative pressure wound therapy with instillation was selected to aid in drawing out any remaining infectious material in the wound, to assist in drawing together the medial and lateral flaps through Ilizarov’s principal, to maintain a moist wound environment, and to bathe the exposed fourth metatarsal head periodically with antibiotic solution to prevent contiguous osteomyelitis and desiccation of the exposed bone.
       On the fifth inpatient day, following a second debridement and washout, NPWT with instillation was continued for 4 days with concomitant intravenous antibiotics and bed rest.
       The dressing was removed on inpatient Day 9 and a healthy, fully granular wound bed with complete coverage of the exposed fourth metatarsal head was noted. All cellulitis and edema had resolved, and the negative pressure had inset the lateral and medial flaps. Negative pressure wound therapy with instillation was discontinued, and the lateral flap was secured to the medial tissue with steri strips. The patient was discharged with a total contact cast with weekly changes until full healing was achieved at 10 weeks.
       After 9 days of NPWT with instillation, the wound was healthy and resolving with weekly total contact cast changes. Without the therapy, the metatarsal head would most likely have been at least partially exposed on Day 9, which is not optimal for patient discharge. Granulation tissue formation over the bone lead to the investigator’s decision for a cast. The cast allowed full weightbearing of the foot and discharge home instead of prolonging the length of stay with transfer to a nursing home, wheelchair use, and daily dressing changes.
       Case study 3: abscess right great toe with osteomyelitis. A 51-year-old White male with type II diabetes mellitus presented in the emergency room with frank abscess of the right hallux and a draining sinus tract extending to the phalanges present for 1 week. Cellulitis to the midfoot was present. Immediate empiric antibiotics via heplock were started, and the wound was incised and drained to decompress the abscess in the emergency department. Twenty-four hours later, the patient underwent operative debridement of the hallux with excision of necrotic bone and pulsed lavage irrigation. Intraoperative cultures revealed Streptococcus agalactiae and broadly sensitive Staphylococcus aureus.
       Negative pressure wound therapy with instillation was initiated on post-operative Day 1. Negative pressure wound therapy with instillation was continued through Day 6 with intravenous antibiotics, bedrest, central line placement, and bedside debridements at each negative pressure dressing change until all signs of clinical infection resolved with granulation at the wound base over bone. The patient was discharged on Day 7 to continue a course of long-term intravenous antibiotics at home with a walker for strict nonweightbearing. Home care nurses provided daily irrigation and packing of the wound. At Week 11, the wound was completely closed.
       Negative pressure wound therapy with instillation stabilized this wound and created a controlled environment for new tissue growth and subsequent coverage of the bone with granulation tissue. The rate of granulation tissue formation allowed the patient to be discharged on Day 7.
       Case study 4: diabetic foot ulcer with osteomyelitis and central compartment abscess. A 54-year-old Black male with type II diabetes mellitus presented with a fever and a left foot ulcer of 3 weeks duration with central compartment abscess, which had developed while at another institution. Examination of the left foot revealed a submetatarsal 4 ulceration with probing into the proximal central compartment (Figures 1A and 1B). During the previous 12 months, the patient had been seeing a podiatrist for regular debridements and dressing changes of saline wet-to-moist dressings, which were not successful in treating the wound. There was a history of first and fifth ray amputations of his left foot. A Doppler examination showed adequate perfusion for healing. Plain radiographs showed dissolution of the fourth metatarsal head. Partial heterotrophic bone formation of the amputated fifth metatarsal was noted distally. Swabs of superficial wound drainage grew Proteus mirabilis while Gram stain revealed heavy gram-positive cocci in pairs and clusters as well as moderate gram-negative rods and slight gram-positive rods. Upon admittance to the hospital, the patient was immediately started on empiric intravenous antibiotics.

Figure 1

       On Day 2 at the hospital, the patient was taken to the operative theater for incision and drainage of the plantar central compartment abscess and excision of the osteomyelitic fourth metatarsal head. Deep intraoperative cultures of Proteus mirabilis mirrored wound swabs at admission. Pathology revealed necrotizing ulceration of skin and subcutaneous tissue and osteomyelitis of cancellous bone. The surgical dressings were removed 2 days later and revealed an extensive cavitating wound with viable peripheral tissue without purulence or cellulitis (Figure 1C). Following the removal of the surgical dressings on inpatient Day 4, NPWT with instillation was initiated with continued intravenous antibiotics and bedrest (Figures 1D and 1E). The dressing was changed on inpatient Day 6, revealing increased granular budding, decreased peripheral wound edema, and contracture of the proximal incision line secondary to the negative pressure (Figure 1F). While NPWT with instillation was continued, review of post-operative radiographs noted severe disruption of the metatarsal parabola, and it was felt that conversion to a transmetatarsal amputation would be necessary to prevent re-ulceration under the remaining 2 metatarsal heads. The patient returned to the operative theater for transmetatarsal amputation on Day 7, re-debridement of the plantar space incision, and a percutaneous Achilles tendon lengthening (Figure 1G). Negative pressure wound therapy with instillation was re-initiated on Day 8.
       On Day 9, the wound was clear of clinical infection with granulation tissue coverage over deep tissue and bone (Figure 1H). The patient was discharged for nonweightbearing with crutches and daily dressings at a skilled nursing facility. Complete healing was noted at 4 1/2 weeks (Figure 1I).
       In the authors’ opinion, the NPWT with instillation converted this large, cavitating, and colonized wound to a granulating and clean wound in 7 days. This dissuaded the authors from moving in the direction of a more proximal amputation, which would increase hospital costs and morbidity for the patient. Secondly, conversion to a clean and granulating wound allowed a transmetatarsal amputation and primary closure in the operating room, rather than bringing the patient back for a third operative debridement and final closure. The authors would not have been comfortable closing the transmetatarsal amputation primarily if the plantar wound still appeared fibrotic with exposed deep tissues and colonization. This would ultimately have increased hospital stay and costs. In addition to this direct cost savings, the authors believe that a more proximal amputation was avoided, which could have resulted in an additional cost of approximately $22,000 to $36,000 depending on the method used.
       Case study 5: recurrent diabetic ulcer with osteomyelitic heterotrophic bone. A 56-year-old west Indian woman with type II diabetes mellitus presented with a recurrent ulceration of 3 weeks duration on the lateral border of her right foot at the site of previous amputation of the fifth toe and metatarsal. Clinical examination revealed a fibrotic ulceration with sinus tract to bone with serous drainage and periwound induration and erythema (Figure 2A). Noninvasive vascular examination and previous operative experience revealed adequate perfusion of the foot. Plain radiographs revealed significant heterotrophic regrowth of the previously resected fifth metatarsal right foot (Figure 2B). The patient was started on empiric intravenous antibiotics and on Day 3, the patient was taken to the operative theater for wide debridement of the sinus tract and underlying infected bone. Intraoperative tissue cultures revealed heavy growth of Escherichia coli and broadly sensitive coagulase-negative staphylococcus with slight Enterococcus fecalis and Staphylococcus aureus. Gram stain revealed moderate gram-negative rods, slight gram-positive cocci in pairs, and moderate polymorphonuclear neutrophils. Tissue pathologic examination showed bone with chronic fibrosing periostitis. The wound was packed open with a compressive surgical over-dressing throughout Day 4 with the white count and signs and symptoms of infection trending down.
       On Day 5, the surgical dressings were removed for wound inspection, revealing that the base of the wound contained healthy bone with exposed metatarsocuboid joint (Figure 2C). Negative pressure wound therapy with instillation was initiated in an effort to bathe the bone and joint with antibiotic solution and develop rapid granulation for osseous and articular coverage (Figure 2D). Negative pressure wound therapy with instillation and intravenous antibiotics were continued through Day 10 with 48-hour interval changes and central venous line placement on Day 9 for long-term intravenous antibiotic delivery.
       On Day 10, NPWT with instillation was removed, and the wound was inspected. Profuse granulation tissue was present throughout the wound base with complete coverage of deep tissue structures (Figure 2E). The patient was discharged to a skilled nursing unit on Day 12 for daily moist dressings, intravenous antibiotics, and wheelchair restriction with complete healing at 2 months (Figure 2F).

Figure 2

       This recurrent infection was surgically treated and promptly granulated through use of the NPWT with instillation. Coverage of this large wound and concomitant bone and joint structures allowed transfer to a skilled nursing facility for simple moist dressings rather than possible additional surgery, such as pedicle or free flap grafting. In the authors’ opinion, plastic surgical procedures or more proximal amputation would have been necessary without NPWT with instillation before transfer due to the extent of the wound and the amount of deep tissue exposure. The patient was able to recover uneventfully in the nursing home without additional surgery and without the morbidity of proximal amputation. The acute care length of stay was limited to 10 days rather than up to 20 days if the additional procedures had been performed.

Discussion

       Animal models have shown that epithelialization occurs twice as fast in a moist environment under a film dressing than when exposed to air.³⁷ Moist wound healing has become the standard of care.³⁸ Uncontrolled build-up of bacterial byproducts as well as chronic wound fluid have been shown to inhibit wound healing.^39–41 Mechanical stress on cells has been shown to stimulate new growth.^42–44 Negative pressure wound therapy addresses all of these issues. The study of wound healing in an antibiotic-laced, fluid-filled chamber has also been documented by Vranckx et al.⁴⁵ This article also contains an excellent review of wet wound healing history over the last 200 years. The advantages that the authors identify with this technique are decreased pain and progression of tissue necrosis, ease of administration of growth factors, analgesics, and antibiotics, decreased scarring, increased ability of cell-to-cell communication with extremely elevated local concentrations of antibiotics, and decreased systemic toxicity secondary to absorption. One can see the similarities between this fluid-chamber model and the NPWT with instillation scenario. While these techniques certainly provide a moist wound environment, one must also provide at least intermittent optimal irrigation pressure to remove debris and adhered bacteria when dealing with heavily contaminated wounds. The optimal irrigation pressure to enhance wound cleansing without causing trauma to the wound bed has been shown to be 4–15 pounds per square inch (psi). Piston and high-pressure syringe systems deliver 4.2–8 psi. Pour bottles and bulb syringes deliver 0–1 psi. High-pressure lavage can deliver between 8.8 psi and 70 psi depending on design and has been shown to increase tissue damage in bacterial penetration at the highest settings.^46–50 As discussed by Wolvos,³⁶ NPWT with instillation is a low-pressure delivery system. Therefore, the authors agree that irrigation at clinically effective pressures (4–15 psi) at the time of operative debridement and at each negative pressure dressing change would be wise. The choice of irrigant should take into consideration a detailed history of allergic drug-induced urticaria, anaphylaxis, antibiotic allergies, and renal status. In addition to the drug characteristics of the particular solution, the surface area of the wound will also determine the amount of systemic uptake. The minimum inhibitory concentration that will be present at the wound level and likely organisms should be considered. One should also consider the amount of exposed bone as a potential systemic access site for drugs. Lastly, regular pharmacokinetic monitoring is mandatory if solutions with known toxicity are being instilled.

Conclusion

       After the authors’ initial experience with NPWT with instillation, a basic list of scenarios that would prompt conversion to this modality has emerged. Negative pressure wound therapy with instillation would be utilized when:
• Diffuse or extensive osteomyelitis is encountered or if large areas of post-debridement bone or joint are present
• A question remains regarding completeness of debridement of infected bone
• Pain secondary to surgical procedures complicated by the standard negative pressure dressing changes is significant enough to warrant low-dose local anesthetic infusion
• A wound is suspected to be critically colonized (increased drainage, color changes to drainage or wound base, increased pain, odor, surface cultures of organisms that are pathologic at low numbers, multispecies cultures) and subsequently has become indolent or “stalled” yet needs continued negative pressure therapy⁵¹
• An alternative to antibiotic-impregnated beads is needed.
       The authors’ practice is to utilize NPWT without instillation in any wound when complete excision of infected bone (eg, disarticulation) can be guaranteed or when the debridement is limited to soft tissue. Combining instillation therapy with the existing negative pressure dressings should help assist in decreasing overall wound fluid viscosity, removing infectious materials, and lowering the bioburden to convert an infected or critically colonized wound to a clean or contaminated wound. The authors have limited the use of the antibiotic bead pouch technique and discontinued instillation with passive egress and closed suction irrigation in favor of the new negative pressure dressing with instillation. The replacement of the antibiotic bead pouch technique with NPWT with instillation has been contemplated by surgeons in other specialties, such as first stage management after open fractures in orthopedic traumatology (Neiman R, personal communication). Antibiotic beads are still utilized in a more limited fashion. Typically, the beads are used when osteomyelitis is present and sinus tract formation is to be excised and primarily closed at the time of bone resection in a preparatory stage prior to final reconstruction. The beads are not utilized when a large wound is present or if a second stage is not contemplated.
       Current use of NPWT with instillation by the authors has continued, but the solution utilized has changed. The combination of bacitracin and polymyxin B solution as previously utilized by the authors does expand the antimicrobial action to both gram-positive and gram-negative bacteria as compared to bacitracin alone. A thorough examination of the aforementioned double antibiotic solution reveals several drawbacks, however. A replacement agent for the double antibiotic solution prompted an exhaustive review of the many topical antimicrobial and antiseptic agents available today, showing that no single ideal solution exists. An ideal agent would have a broad spectrum of activity with no systemic absorption. In addition, the ideal agent would be painless, inexpensive, and nontoxic with minimal wound healing interference. Lastly, this agent would be approved for topical use in solution form by the US Food and Drug Administration (FDA).
       While all surgeons are familiar with the use of topical antibiotics for wound irrigation, most are aware of the lack of proven efficacy. Many of the typical antibiotics utilized as wound irrigants have not been approved for preparation and use in solution by the FDA despite their widespread use. This list includes vancomycin, polymyxin, neomycin, cephalosporins, kanamycin, gentamycin, and bacitracin. Moreover, some drugs on this list can have potent nephrotoxic and ototoxic side affects, while others have been associated with dangerous resistance associated with troughing that occurs when used as an irrigant. Other solutions, such as acetic acid, alcohol-based solutions, Dakin’s solution, povidone iodine, and hydrogen peroxide, are either substantially cytotoxic or degrade components of the negative pressure dressing and should be avoided. Povidone iodine has been documented as the cause of death after irrigation,⁵² while hydrogen peroxide has lead to embolization and coma.⁵³ With specific attention to the authors’ combined use of bacitracin and polymyxin B, the following items should be considered. Bacitracin has been shown to interact with neuromuscular blocking agents. A study on patients receiving total hip replacements showed an increase in nonfatal post-operative renal failure incidence when patients had bacitracin-neomycin-polymyxin B solution wound irrigation.⁵⁴ This is of particular importance in the diabetic population with the expected increased incidence of renal insufficiency. Anaphylaxis has been documented with the use of antibiotics for topical irrigation, including bacitracin.⁵⁵
       In addition, choices among the short list of antibiotics that are FDA approved for topical use in solution form can have potential drawbacks as well. Mafenide acetate is FDA approved for use in a 5% solution form as a topical agent. Mafenide acetate is effective in the presence of pus or an acidic environment. This agent is broadly bacteriostatic against many gram-positive and gram-negative bacteria as well as some anaerobes. However, this solution has been shown to be systemically absorbed. Fatal hemolytic anemia with disseminated intravascular coagulation has been reported. Metabolic acidosis is a known complication of topical use of mafenide acetate, and care should be taken when utilizing this solution in patients with pulmonary or renal dysfunction. Rare bone marrow depression and poryphyria cases have been reported. Also of importance are the increase in local rash and pruritus and the almost univeral pain and burning sensation expected with its use. Cross sensitivity in patients allergic to other sulfonamides is possible. Lastly, cytotoxicity to human keratinocytes has been shown in vitro.
       Silver nitrate 0.5% solution is also approved by the FDA and is bactericidal against panresistant bacteria. When using NPWT with instillation, the authors are contemplating the use of silver nitrate solution rather than the topical antibiotics detailed in the 5 case studies. Silver nitrate has a proven track record at many burn centers throughout the US for the decolonization of burn wounds prior to skin grafting. The solution is painless when applied to the wound and is not absorbed into the blood stream, which has been demonstrated through serum level tests. This solution does not damage the components of the negative pressure dressing. The silver nitrate solution is prepared from the lyophilized powder, which is reconstituted with sterile water. Bacterial resistance can develop to silver nitrate solution but is distinctly uncommon. The solution is hypotonic and therefore not an ideal environment for healing tissue. However, the therapy should be seen as a bridge rather than a therapy to closure. A continuous flow gravity feed at 5 cc/hour with the vacuum unit set at 125 mmHg continuous has been proposed. The theory on this application is continuous flow over the wound bed (Bauer G, personal communication). It is important that the pharmacist prepare the solution carefully, as the 0.5% solution has minimal cytotoxicity with an optimal kill rate of bacteria. Solutions greater than 0.5% can denature proteins. Methemoglobinemia is a possible complication at higher concentrations due to the reduction of nitrate to nitrite causing oxidant-induced tissue damage.⁵⁶ Silver nitrate at a concentration of 5% to 10% produces chemical cautery. Minor complications of the use of silver nitrate include staining of bed linens and instability to light. Based on these facts, the authors have changed the instillation protocol to include silver nitrate solution rather than topical administration of antibiotics. Lastly, the addition of 25 cc of 1% lidocaine in the 250 cc bag of irrigant will be considered in the rare occasion that pain is a significant entity in the authors’ patient population. Silver nitrate 0.5% solution has been proposed as an optimal solution for use with NPWT with instillation. The authors’ experience has shown that NPWT with instillation has much promise as an adjunct to systemic antibiotics and will aid in withdrawing any remaining infectious material in the wound, enhancing granulation tissue formation, maintaining a moist wound environment, and periodically bathing the wound with antibiotic solution to prevent contiguous osteomyelitis as well as desiccation of the exposed bone.

Future Studies

       A statistically powered study with quantitative tissue biopsies at each dressing change of subatmospheric dressing versus the same dressing with the addition of topical antimicrobial solution in a stratified population would give objective evidence supporting this new device.