Efficacy of hydrosurgical debridement and nanocrystalline silver dressings for infection prevention in type II and III open injuries

Abstract

The aim of this study was to retrospectively evaluate the clinical and culture‐positive infection rates of open Gustilo/Anderson type II and III fractures using a protocol nanocrystalline silver wound dressing and hydrosurgical debridement. Retrospective case series through chart review on all type II and III open fractures were treated using a novel protocol from December 2005 to March 2008 (N = 17). All Gustilo/Anderson grade II and III open fractures were treated with a novel protocol at a Level I trauma centre. Open Gustilo/Anderson grade II and III fractures were acutely stabilised in the trauma centre/emergency department, while a nanocrystalline silver dressing was placed within the wound. Debridement using a hydrosurgical scalpel and gravity irrigation was performed within 6–8 hours of injury. Cultures were obtained prior to definitive fixation. The primary outcome measurements were positive cultures and clinical infection rates. Seventeen patients met inclusion criteria. Mean age (33·5) and injury severity score (12·7) were gathered. There were 4 grade II open fractures (23·5%), 11 grade IIIA (64·7%) and 2 grade IIIB open fractures (11·8%). The mean time to intravenous antibiotics was 61·5 minutes. The mean time to initial debridement/irrigation was 222·1 minutes. The average number of surgical procedures was 2·35 with a mean length of stay of 11·8 days. Six patients developed positive cultures from the traumatic wounds, five were contaminants. One clinical infection was found (methicillin‐resistant Staphylococcus aureus). The overall clinical infection rate in this series was 5·9% (1/17). The only infection was in a Gustilo/Anderson grade II fracture. There were no infections in the more high‐energy Gustilo/Anderson grade IIIA and IIIB fractures compared with the Gustilo/Anderson control of 4–42%. We conclude that this novel protocol for open‐fracture treatment is a promising intervention. A further prospective randomised clinical study is warranted.

INTRODUCTION

Open fractures present a unique challenge for the treating surgeon as two critical issues must be addressed. Bacterial contamination at the fracture site most commonly originates from the skin and environment and presents a major barrier to healing. Disruption of the surrounding soft‐tissue envelope often threatens vascularity to the underlying bone. Open fractures are therefore at great risk of infection and delayed union particularly in the setting of high‐energy trauma.

Early antibiotic prophylaxis followed by urgent debridement and irrigation (D & I) and stabilisation are the gold standard for prevention of infection of open fractures 1, 2. Multiple D & I's are often necessary prior to definitive surgical stabilisation, prolonging time to recovery and hospital stay and generating a considerable cost to the health care system. A major factor in achieving fracture healing is maintaining or providing a viable soft‐tissue envelope for blood supply. Debridement should therefore proceed in a balance between excision of all non viable and potentially septogenic tissue versus iatrogenic disruption of the soft‐tissue envelope. For over 20 years water‐jet dissection has been described as an effective and safe alternative surgical tool for liver and brain surgery 3, 4. A water‐jet functions by pushing water through a nozzle of various diameters under a pressure of 1–150 bars (5). A recently developed hydrosurgical debridement tool generates a water‐jet across a 1/2‐inch aperture located at the tip of a hydrosurgical scalpel (Versajet®, Smith and Nephew, Memphis, TN). On the basis of the Venturi effect/Bernoulli principle, the flow of this jet across the operating window generates a vacuum, thereby incising and removing injured or contaminated tissue.

Silver has been used in modern medicine as an effective antimicrobial agent for nearly two centuries 6, 7. Silver has been shown to be cytotoxic for keratinocytes and fibroblasts in vitro (8). Cytotoxicity of silver depends on the method of application, with a higher toxicity found in silver sulphadiazine, compared with nanocrystalline silver dressings (9). In vivo studies on full thickness skin wounds have shown no differences of percentage of cultured skin substitute engraftment in the presence of silver, saline or an antibiotic ointment (10). Clinically, silver has been used mainly as a liquid (silver nitrate) or incorporated in cream (silver sulphadiazine) for the management of burn wounds and the prevention of associated burn sepsis. The major drawback of these methods of application is fast silver deactivation by serum elements, making a more frequent application necessary (11). A nanocrystalline silver dressing has been developed to prevent wound adhesion, limit nosocomial infection, control bacterial growth and facilitate burn wound care (Acticoat®, Smith and Nephew) (12). The theoretical advantage of an increased silver surface area has been clinically shown to provide fast and sustained antimicrobial activity for up to 7 days against a broad spectrum of bacteria including methicillin‐resistant Staphylococcus aureus (MRSA) and vancomycin‐resistant Enterococcus (VRE) that is both bacteriostatic and bactericidal 13, 14.

Hydrosurgical debridement and nanocrystalline silver dressings have separately been used with promising results in different arenas for the management of necrotic and burn wounds (3) but have not been previously described as an approach to manage open fractures. We hypothesise that treatment of open fractures using nanocrystalline silver dressings and hydrosurgical debridement decreases clinical infection rates compared with historical controls.

METHODS

After obtaining Institutional Review Board approval, a retrospective case series was conducted from December 2005 to March 2008 to evaluate the efficacy and safety of hydrosurgical debridement and concomitant application of nanocrystalline silver wound dressings in the prevention of infection of open Gustilo type II and III fractures 15, 16. Inclusion criteria were age greater than or equal to 18 years, all patients with Gustilo type II and III open fractures treated with hydrosurgical debridement and nanocrystalline silver dressings and existing culture results. Exclusion criteria include paediatric patients, incomplete medical records and open fractures not treated according to protocol, history of recent infection in the injured extremity, extremities requiring primary amputation, systemic antibiotic use 2 weeks prior to fracture, immunosuppresant use with 1 month of fracture and allergy to silver. The following baseline parameters were gathered and reported by the way of chart review and trauma registry search: age, gender, comorbidities, fracture site, open‐fracture grade, fracture type (AO classification), Injury Severity Score (ISS), prophylactic antibiotics used, time to initial antibiotic dose, time to D & I and species of culture‐positive bacteria (Table 1). Outcome measures include laboratory culture results, presence of clinical infection, number of D & I procedures prior to definitive fixation, length of hospital stay (LOS) and fracture healing. Wound cultures taken at the last D & I prior to installing definitive hardware serve as the results reported. All radiographs were reviewed and healing was defined as follows: bridging callus on at least three cortices as assessed on orthogonal plain radiographs and weight bearing without pain at the fracture site.

Table 1.

Baseline characteristics and patient demographics

Patient	Age	Gender	Comorbidity	Smoker	Open injury	Classification	ISS	Time to antibiotics (min)	Antibiotics	Time to OR (min)	Positive culture	Organism	Procedures	LOS	Complications	Injury healed
1	29	Female	Bipolar	Yes	Tibia (pilon)	IIIb	14	40	Ancef and Gentamycin	180			3	4	STSG	Yes
2	19	Male		Yes	Patella	II	12	180	Ancef	180	Yes	MRSA	1	2	Clinical infection	Yes
3	34	Male			Elbow arthrotomy	II	14	40	Ancef	180			1	7		Yes
4	41	Male			Femur	II	22	60	Ancef	240			1	30		Yes
5	19	Male			Knee arthrotomy	IIIa	5	15	Ancef and Gentamycin	240			2	5		Yes
6	19	Male			Femur	II	13	30	Ancef	270			1	30		Yes
7	23	Female		Yes	Tibia (ankle)	IIIa	14	30	Clindamycin	320	Yes	MSSA	3	14	ROH	Yes
8	12	Male			Tibia, olecranon	IIIa	22	120	Ancef	180			3	10		LFU
9	59	Male	Hypertension, asthma	Yes	Tibia (pilon)	IIIa	9	35	Ancef	295			3	5		Yes
10	26	Male	Alcohol		Tibia	IIIa	9	10	Clindamycin	105			1	7		LFU
11	47	Male		Yes	Tibia	IIIa	13	40	Ancef and Gentamycin	180			4	16		Yes
12	28	Male		Yes	Tibia	IIIb	9	10	Ancef and Gentamycin	180	Yes	Pseudomonas	2	25		Yes
13	19	Male			Olecranon, ulna	IIIa	22	90	Ancef and Gentamycin	360	Yes	Enterobacter	3	16
14	55	Male			Tibia (pilon)	IIIa	10	130	Ancef	135	Yes	Coagulase‐negative Staphylococcus	4	11	ROH	Yes
15	49	Female			Bilateral pilon	IIIa	9	10	Ancef	245			3	5		Yes
16	52	Female	Diabetes, hypertension, heart disease, hyperlipidemia		Tibia (ankle)	IIIa	10	175	Ancef	248			10	10		Yes
17	39	Female			Subtalar dislocation	IIIa	9	30	Ancef	238	Yes	Serratia marcescens	4	4		Yes
Mean	33.5						13	61.5		221.1			2.35	11.8

* Classification, Gustilo/Anderson classification; ISS, Injury Severity Score; LFU, lost to follow‐up; LOS, length of hospital stay (days); min, minutes; MRSA, methicillin‐resistant Staphylococcus aureus; MSSA, methicillin‐sensitive S. aureus; ROH, removal of hardware; STSG, split‐thickness skin graft.

Treatment protocol

The following treatment protocol was designed and implemented by a fellowship‐trained orthopaedic trauma surgeon, the senior author (MS). All procedures were performed by the senior author (MS). Open fractures were acutely irrigated in the trauma centre/ emergency department while receiving intravenous antibiotics, and a nanocrystalline silver dressing was placed within the wound while plaster immobilisation was applied. Urgent hydrosurgical debridement and gravity irrigation of bone and soft tissue within 6–8 hours of injury follows. At this time a decision is made regarding definitive wound closure. If the wound requires additional D & I's, a fresh nanocrystalline silver dressing is placed within the wound with an overlying negative‐pressure dressing (Figure 1A, B). A repeat D & I is then performed 48–72 hours after the injury. Wound cultures are taken at the end of each procedure. This procedure is repeated until negative cultures are obtained. Definitive fixation is performed during the subsequent surgery and the silver dressing is then placed purely as a postoperative dressing.

Figure 1.

Treatment protocol – nanocrystalline silver dressing (A) before and (B) after wound packing.

RESULTS

A total of 17 patients, 5 females and 12 males, met inclusion criteria. The average age was 33·5 (SD = 14·8, range 12–59), ISS was 12·7 (SD = 5·04, range of 5–22). There were 4 grade II open fractures (23·5%), 11 grade IIIA fractures (64·7%) and 2 grade IIIB open fractures (11·8%). All received parenternal antibiotics. Seven patients were smokers, one patient was diabetic. The mean time from arrival in the trauma centre to antibiotic administration was 61·5 minutes (SD = 56·5 minutes, range 10–180). Five patients received antibiotics after 60 minutes, while three patients received antibiotics after 120 minutes. Ten patients received cephalexin only, five patients received cephalexin and gentamycin and two patients received another combination of antibiotics due to penicillin allergy. Mean time to initial D & I was 222·1 minutes (SD = 66 minutes, range 105–360). The average number of surgical procedures was 2·35 (SD = 1·05, range 1–4). The mean LOS was 11·8 days (SD = 8·9, range 2–30). Of the 17 patients included in the study, six developed positive cultures from their wounds. Five were deemed contaminants (coagulase‐negative S. aureus, Enterobacter, Pseudomonas, Serratia) due to the absence of clinical infection. The remaining culture‐positive case was the only clinical infection in the series. It involved a grade II open patella fracture treated as per the experimental protocol and subsequently underwent partial patellectomy with patella tendon repair. The patient went on to develop an MRSA infection at the surgical site 2 weeks after his injury. For this infection the patient underwent four more D & I's and negative‐pressure therapy on a subsequent admission, followed by 4 weeks of organism‐specific parenteral antibiotic treatment. Patients were followed until healing was documented. Healing was documented in 14 of the 17 patients, while 3 patients were lost to follow‐up (follow‐up rate = 82%). Complications include painful hardware requiring removal (two), clinical signs of infection (one) and the need for split‐thickness skin graft (one) (Table 1).

DISCUSSION

The association between open fractures, soft‐tissue disruption, infection and non union is well documented 17, 18, 19, 20. Esterhai et al. underlined the importance of the soft‐tissue envelope to fracture healing (21), while Fischer et al. chronicled the importance of early and adequate debridement to allow for appropriate soft‐tissue coverage within the first 2 weeks (22). Meticulous and complete debridement of the bone and soft tissues is a critical step towards union.

Recent work performed by Lenarz et al. showed success with the timing of wound closure in open fractures based on cultures obtained after debridement (23). Their protocol consisted of obtaining aerobic and anaerobic cultures after the initial debridement and irrigation. At 48 hours after debridement, patients returned to surgery for repeat D & I and cultures. If initial cultures were positive, the wound was left open and the patient returned to the OR in 48 hours. This procedure was repeated, and the wound was not closed until negative culture results were achieved (23).

Clearly, there is a premium on the quality of the D & I in infection prevention. However, questions have been raised regarding the methods used to perform D & I's. Adili et al. claim that high‐pressure pulsatile lavage is deleterious to the mechanical strength of early fracture callus (24). Furthermore, others have reported that irrigation and debridement can be a source of iatrogenic injury in open‐fracture treatment as it disrupts soft tissues at a deeper level than low‐pressure lavage (25). Moreover, Park et al. have shown that repeat D & I's associated with persistent rigid immobilisation may contribute to delayed unions or atrophic non unions (26).

A recent international survey of 984 surgeons inquired their preference in fluid lavage in patients with open‐fracture wounds (FLOW 2008) (27). Conclusions drawn by Petrisor et al. were that the majority of surgeons favour both normal saline and low‐pressure lavage for the initial management of open‐fracture wounds. However, opinions varied with regard to the efficacy of different solutions, the use of additives and high versus low pressure (27).

Other treatment adjuncts in the management of open fractures include the use of negative‐pressure wound therapy (NPWT) and reticulated open cell foam (ROCF) (28). Webb et al. described two main theories regarding the mechanism of action of NPWT/ROCF. The first is based on the stimulatory effect of microstrain on cellular mitogenesis, angiogenesis and the elaboration of growth factors. This is much like the controlled Ilizarovian distraction or tissue expansion. The second is based on the enhancement of the dynamics of microcirculation by active evacuation of excess interstitial fluid in the form of oedema. Physiologically, there is a lowering of the heightened capillary afterload and a qualitative dilution of contained microcontaminants, bacteria and proinflammatory cytokines (28).

Stannard et al. performed a prospective randomised study comparing the treatment of open fractures with negative wound pressure therapy versus a control (29). The use of NPWT resulted in one fifth the infection rate as did the control group. Their conclusion was that NPWT represents a promising new therapy for severe open fractures after high‐energy trauma (29).

The use of silver‐impregnated ROCF has been studied by Schlatterer et al. (30). Much of the available information comes from the manufacturer and there is a need for randomised clinical trials using silver‐impregnated NPWT. For example, depth of penetration of the silver into the wound is an elusive question because depth implies delivery and the dressings are not delivery devices. One side effect noted has been Argyria of discoloration of the skin/wound bed. The distinction for the surgeon treating traumatic wounds is that the darker tissue colour in the wound after use of the sliver foam dressing may mislead one to suspect further tissue necrosis. However, in the acute, highly traumatised wound, which is yet to declare itself, silver‐impregnated NPWT may have a role. This type of wound bed theoretically is less well perfused in the central region of the wound. The silver foam dressing can be advantageous with the intent of infection prophylaxis (30).

The 2010 Cochrane Review, ‘Topical silver for preventing wound infection’ addresses the outcomes of 26 clinical research trials evaluating the usage of silver‐containing wound dressings and topical agents in the healing of wounds and the prevention of wound infections (31). Two particular studies compared nanocrystalline silver‐coated dressings (Acticoat®) with non silver dressings in subjects who suffered from burns. Ultimately, the Cochrane Review concluded that more evidence is necessary to establish whether silver‐containing dressings and topical agents contribute to wound healing and/or the prevention of wound infections (31).

As referenced in the 2010 Cochrane review, ‘Topical silver for preventing wound infection (Review)’, Issue 3, Nanocrystalline silver‐coated dressing (Acticoat®) was compared with hydrophilic polyurethane dressing (Allevyn®) (31). Innes evaluated and enrolled 17 patients with 18 paired adjacent burn sites, who required a split‐thickness skin graft. His findings showed no statistically significant difference in the number of patients who developed an infection or the number of positive cultures at any time point. Healing was significantly accelerated in the hydrophilic polyurethane dressing group (Allevyn®) at 9·1 days versus the nanocrystalline silver‐coated dressing (Acticoat®) group at 14·5 days.

Tredget (12) compared nanocrystalline silver‐coated dressing (Acticoat®) with fine‐meshgauze with silver nitrate (0·5%). The study showed that the nonocrystalline silver‐coated (Acticoat®) group developed significantly fewer wound infections (5/17) when compared with the fine‐mesh gauze with silver nitrate (0·5%) group (16/17). Regarding healing rates, the authors stated that there was no difference when comparing the two groups; however, no data were reported that supported this claim (31).

Using the protocol described above for open‐fracture management, our retrospective case series produced an overall clinical infection rate of 5·9% [95% confidence interval (CI) = 0·001–0·29]. The sole clinical infection was a type II open fracture, accounting for 25% of all type II open fractures (1/4 patients, 95% CI = 0·006–0·8). This patient did not receive antibiotics until arrival in the operating room, 180 minutes after the injury. Although our sample size is too small to validly assess the relationship between open‐fracture grade and infection, it is noteworthy that no patients with type III fractures developed clinical infections compared with historical values of 4–52% 15, 16, 32, 33. No significant association could be observed between positive culture results or clinical infection and length of hospital stay; possibly because many patients in this series sustained polytrauma. No cases of VRE were observed, while the sole infection was due to MRSA.

In summary, although purely a case‐series involving multiple interventions, we report encouraging early results using our protocol for open fractures. The benefit of infection reduction has the potential to reduce operative procedures, morbidity and cost, while improving outcomes. These findings may be of particular significance in the setting of high energy, contaminated open fractures (Gustilo/Anderson type III) for which no clinical infections were observed. Further randomised, prospective study is warranted.