The Effect of a Bacteria- and Fungi-binding Mesh Dressing on the Bacterial Load of Pressure Ulcers Treated With Negative Pressure Wound Therapy: A Pilot Study

Objective. This study was designed to clinically evaluate the efficacy of a bacteria- and-fungi-binding mesh (BFBM) dressing to modify the bacterial load of pressure ulcers (PUs) of categories 3 and 4, when used as a wound contact layer (WCL) during negative pressure wound therapy (NPWT). Methods. This was an observational single-centre study in patients with PUs of categories 3 or 4, who were treated with NPWT. Patients were observed for 7 days and received NPWT at -80 mm Hg with the BFBM dressing as the WCL. Wound biopsies were performed at inclusion (B0), at 48 hours (B1), and at day 7 (B7). Bacteria- and fungi-binding mesh dressings were examined for bacterial load at 48 hours (D1) and at 7 days (D7). The primary endpoint was the changes in bacterial loads. Results. Fifty patients were enrolled; 43 (86%) of their PUs were on the sacrum. At B0, 3 groups of wounds were identified by the bioburden level: group A had negative results (28%) to bacterial loads from 10² to 5 x 10³ colony forming units (CFU) CFU/mL (18%); group B had 10⁴ to 10⁵ CFU/mL (18%); and group C with ≥ 106 CFU/mL (36%). The authors did not find any significant difference in bacterial loads in group A, but significant differences were found in group B at B1 and B7 (P = 0.04 and P = 0.0067) and in group C at B1 and B7 (P < 0.00001). There was no significant difference on the bacterial loads of the dressing at D1 and D7 (P = 0.823). No device-related adverse events were reported. Conclusion. The BFBM dressing seems to be at the origin of a statistically significant reduction of bacterial burden in wounds with moderate or high levels of colonization. The authors’ findings suggest BFBM dressings may be a WCL of choice during the treatment of chronic wounds with NPWT.

A pressure ulcer (PU) is a localized injury to the skin and/or underlying tissue usually over a bony prominence, as a result of pressure, or pressure in combination with shear forces and/or friction. Studies developed by the Statistical Organization on Italian Healthcare (SIC) established that approximately 2 million people suffer from PUs; this number may increase as the aging population increases. Prevention is certainly the best approach in PU management; however, once these chronic wounds are present they represent an increased burden for patients and health care systems. Thus, it is essential to find cost-effective ways to manage the most severe categories of PUs.

Chronic wounds are colonized by a polymicrobial flora. The role of bacteria in wounds depends on their concentration, species composition, and host response. Low concentrations of microbes are considered normal and are not believed to inhibit healing; however, critical colonization and infection are associated with a significant delay in wound healing.¹

The clinical benefits of negative pressure wound therapy (NPWT) are now well recognized. Negative pressure wound therapy has become a common method for treating wounds of different etiologies.² In the past 10 years, published articles studying the role of NPWT on the bioburden of chronic wounds have provided contradictory results.^2-5 

It is understood that dressings may influence the effects of NPWT on wound healing. Foam and gauze are the most frequently used dressings for NPWT. Glass and Nanchahal⁶ published the results of a literature review on NPWT, whereby they concluded from available evidence foam and gauze transmit negative pressure efficiently, there is no clear evidence to favor either dressing, or there is insufficient evidence to credit NPWT with reduced bacterial wound colonization. 

Malmsjö et al⁷ published the results of an experimental study in pigs comparing the performance of 3 different types of dressings: foam, gauze, and a bacteria- and fungi-binding mesh (BFBM) dressing in NPWT with regards to pressure transmission, fluid retention, wound contraction, and microvascular blood flow. Based on the findings and pathogen-binding properties of the BFBM dressing, it was concluded a BFBM dressing may be an interesting alternative dressing in NPWT.

The BFBM dressing is a CE (European Conformity) marked wound dressing that irreversibly binds bacteria without the use of any active chemical agent. The BFBM dressing is coated with dialkyl carbamoyl chloride (DACC), a derivative of fatty acids that makes the dressing hydrophobic. The mode of action is based upon hydrophobic interactions developed between the bacterial surface and the coating of the dressing fibers. Hydrophobic substances and associated microorganisms are surrounded by water molecules which encourage their bonding together. As the antimicrobial properties of the BFBM dressing are based on a physical effect, no bacterial resistance is expected to appear.⁸ Previously, the authors performed a study in 30 patients with PUs of categories 3 and 4 treated with NPWT to evaluate the effect on bacterial loads of a polyhexamethylene biguanide (PHMB)-containing dressing (Kerlix, Covidien, Dublin, Ireland); the authors did not observe any bacterial load modification in heavily colonized wounds.⁹

Given the interesting properties of the BFBM dressing, the limited clinical evidence to support its benefits when used in combination with NPWT, and the results of the authors’ previous study,⁹ it was decided to perform a pilot clinical evaluation to explore the efficacy of the dressing in reducing wound bacterial loads and its safety as a wound contact layer (WCL) in patients with PUs of categories 3 or 4 being treated with NPWT.

Two main objectives of this research were: 1) to clinically evaluate the efficacy of BFBM dressings in modifying the bacterial load of PU categories 3 and 4 when used as a contact layer in NPWT, and 2) to evaluate the safety of BFBM dressings through the analysis of any dressing-related adverse events.

This evaluation was made possible by close collaboration between two departments of the Azienda Sanitaria  Locale (ASL) Napoli 3 Sud: the Tissue Repair Centre and the Microbiology Department of the San Leonardo Hospital in Gragnano, Italy. 

Fifty patients in home care (29 women and 21 men, all older than 65) suffering from category 3/4 PUs for 4 or more months, as defined by EPUAP/NPUAP¹⁰ and considered suitable for NPWT treatment after careful clinical assessment, were enrolled in the study. Negative pressure wound therapy is used as a first-line treatment in managing PUs of categories 3 and 4 in Italian health care centers. The authors’ standard protocol includes the application of NPWT during 21 days using PHMB-impregnated gauze as a wound filler and BFBM dressing as the WCL at -80 mm Hg of pressure. The dressing was changed every 48 hours. This protocol of care was followed in the study.

Ethical approval was obtained as well as signed informed consent from patients or their families. Ethical approval was also obtained for the study by the Ethics Committee of the health care district ASL Napoli 3 Sud.

No patients showed clinical signs of local wound infection or systemic infection nor had they received topical or systemic antibiotics within the 7 days prior to commencement of the study. Patients with generally compromised health and those with neoplastic disease or with necrosis in the wound bed were excluded. 

As the study design restricted the observation period to 7 days, it was considered unnecessary for this evaluation to collect data about the wounds’ evolution.

Wound bed biopsies were performed at inclusion (B0) before the application of NPWT to obtain reference values, after 48 hours (B1), during the first dressing change, and at the end of the observation period after 7 days of treatment (B7). Biopsies were taken from the middle of the wound beds, always by the same operator, using a 4-mm punch with plunger system, 7 mm deep (KAI Medical, Honolulu, HI). The authors did not use any local anaesthetic. Samples were placed in sterile tubes without any growth media before being placed in isothermal bags, and they reached the laboratory within 15 minutes for immediate processing. Additionally, samples of the BFBM dressing used were taken and examined for bacterial load at 48 hours (D1) and after 7 days of treatment (D7). The whole dressing was delivered to the laboratory, and a 10% sample corresponding to the central area of each dressing was processed. 

The biopsy and the BFBM dressing samples were processed as follows: 1 mL of sterile physiological saline solution was added to every sample before vortex spinning until it became cloudy (dreggy). The next step was inoculation of 0.01 mL of this solution, using a sterile calibrated loop, onto various microbiological media including standard agar, Columbia CNA Agar for selective growth of gram-positive organisms, Mannitol salt agar (MSA) for selective growth of pathogenic Staphylococci, MacConkey agar for selective growth of gram-negative organisms, and Sabouraud agar for selective growth of fungi. All inoculated media were incubated at 37ºC during 24-48 hours, at aerobic and partial anaerobic conditions. 

After incubation, the total bacterial load, expressed as colony-forming units/mL (CFU/mL) in the wound biopsy or gauze samples, was calculated using the following equation: 

* Counting of visible colonies  

** 1 mL of sterile saline solution was added to the sample

*** 0.01 mL inoculated onto the media

The microorganisms were isolated and identified using bacterioscopic and biochemical methods, including catalase, coagulase, and oxidase activity analysis. Additionally, further identification of the isolated microorganisms was carried out using a Vitek 2 automated system (bioMérieux, Durham, NC).

Recorded information for each patient included data regarding numbers and species for pathogen bacteria and adverse events if present. Numbers and species for saprophyte bacteria were not reported.

Statistical assessment was performed using analysis of variance (ANOVA). Statistical significance was considered at P < 0.05. Data were also analysed using plot graphics to show bacterial load evolution over time.

Among the recruited patients, PU categories 3 or 4 were present in the sacrum of 43 patients, in the trochanter of 3 patients, in the tibial crest, foot, buttock, and above the knee of 1 patient each.

An initial look into the results of B0 bacterial loads showed 4 different categories of wounds: 1) a first group including 14 patients presenting wounds with negative biopsies; 2) a second group including 9 patients presenting wounds with bacterial loads between 10² and 5 x 10³ CFU/mL; 3) a third group including 9 patients presenting wounds with bacterial loads between 10⁴ and 10⁵ CFU/mL; and 4) a fourth group of 18 patients presenting wounds with bacterial loads ≥ 10⁶ CFU/mL.

Bendy et al¹¹ and Noyes et al¹² concluded that bacterial loads ≥ 10⁶/g of tissue or mL of fluid prevent PU healing and can cause invasive infection. The authors used these results to support their decision of reporting any bacterial load ≥ 1 million CFU as ≥ 1 000 000 to facilitate recording and data analysis.

Analysis of variance (ANOVA) did not show any significant difference in bacterial loads in group 1 for bacterial loads in biopsies at B0, B1, and B7 (P = 0.1745); group 2 for bacterial loads in biopsies at B0, B1, and B7 (P = 0.2789); and the results of groups 1 and 2 combined in biopsies at B0, B1, and B7 (P = 0.3022).

On the contrary, ANOVA analysis showed a statistically significant reduction of bacterial loads in biopsies of group 3 at B1 (P = 0.04) and a more important reduction at B7 (P = 0.0069), and group 4 at B1 and B7 (P < 0.00001).

When considering the full range of bacterial load data from biopsies (in all patients),  ANOVA confirmed a statistically significant reduction (P = 0.000016).

Regarding the number of bacteria found in the BFBM dressings,  ANOVA did not show any significant difference in any of the groups at D1 or D7 (P = 0.823).

Taking the results into consideration, the authors considered it reasonable to classify patients by wound bioburden level as follows: group A (n = 23) with a bacterial load ≤ 10³ CFU/mL (low colonization level); group B (n = 9) with a bacterial load ≥ 10⁴ and ≤ 10⁵ CFU/mL (moderate colonization level clinically equivalent to critical colonization); and group C (n = 18), with a bacterial load ≥ 10⁶ CFU/mL (high colonization level clinically equivalent to local/invasive infection). Samples B1 and B7 were not received for 2 patients in group A, so B0 results were not used in any further analysis.

Wound biopsies were performed at inclusion (B0), at 48 hours (B1), and at day 7 (B7). Bacteria- and fungi-binding mesh dressings were examined for bacterial load at 48 hours (D1) and 7 days (D7).

The prevalent organisms isolated in the B0 biopsy samples were Pseudomonas aeruginosa (30%), Escherichia coli (26%), Staphylococcus aureus (13%), and Proteus spp (11%).

In group A (n = 23) (Table 2), all wounds demonstrated low bacterial loads at B0. Bacterial count increased in only 2 patients, with PUs in the sacral area, at B1 and remained high at B7. Isolated bacteria for these 2 patients were: P.  aeruginosa in 1, and a combination of gram-negative bacteria (including E. coli, P. aeruginosa, and Serratia marcescens) in the other. 

Table 1A, 1B presents the bacterial load results for each patient.  Any value > 1 million CFU was reported ≥ 1 000 000 based on the original work by Bendy et al¹¹ and Noyes et al.¹² 

Table 2 shows the results of the bacterial speciation analysis at B0 and B7, comparing the biopsy samples with the corresponding wound dressing samples at D7 from patients in group A, who had wounds at B0 with a bacterial load ≤ 10³ CFU/mL (low colonization level). Each line in the Table 2 represents an individual wound assessed. 

In group B (n = 9), the majority of initial bacterial loads were on the 10⁵ limit (Table 3). They all decreased except in 2 patients at B1 and remained high in only 1 patient with a PU in the sacral area. Results at B7 for this patient showed a combination of E. coli and Proteus spp. 

Table 3 presents the results of bacterial speciation analysis at B0 and B7, comparing the biopsy samples with the corresponding wound dressing samples at D7 from patients in group B, who had wounds at B0 with a bacterial load ≥ 10⁴ and ≤ 10⁵ CFU/mL (moderate colonization level clinically equivalent to critical colonization).  Each line in the Table 3 represents an individual wound assessed. 

In group C (n = 18), all initial bacterial loads (B0) were in the ≥ 10⁶ range (Table 4). They all decreased at B1 except in 2 patients and remained high in only 1 patient at B7. This last patient had a sacral PU; the responsible bacterium was E. coli. 

The results of bacterial speciation analysis at B0 and B7, comparing the biopsy samples with the corresponding wound dressing samples at D7 from patients in group C, who had wounds at B0 with a bacterial load ≥ 10⁶ CFU/mL (high colonization level clinically equivalent to local/invasive infection) appear in Table 4. Each line in the Table 4 represents an individual wound assessed. 

The bacterial loads of all isolated bacterial species from biopsies, whether gram positive or gram negative showed a significant reduction or no increase in 92% of patients. Bacterial load response to treatment is presented in Figures 1, 2, and 3.

Across the 3 groups, bacterial loads in D1 and D7 samples where high even when the wound biopsy results (B1 and B7) were negative (Figure 4). 

The correlation between the bacterial species isolated from wound biopsies and the bacterial species found in the dressings were 77% in group B and 100% in group C. In group A, this correlation could not be established as the B0 biopsies provided negative bacterial loads in 12 patients. However, it should be noted that even if biopsies were negative, the cultures from dressing samples were positive with consistently high levels of bacterial numbers in all patients.

One patient died during the evaluation as the result of complications due to comorbidities. There were no adverse events of any kind related to the BFBM dressing during the evaluation period.

Evidence-based recommendations from the International Expert Group on Negative Pressure Wound Therapy,² published in 2009, establish that antimicrobial gauze dressings may contribute towards infection control. This recommendation was based on studies performed with antimicrobial dressings but in the absence of NPWT. 

The authors started using NPWT for the management of PUs 5 years ago, and almost immediately found out what happened to the bacterial loads when treated with antimicrobial dressings under NPWT. Given the absence of evidence on this subject, the authors initially evaluated the effect on bacterial load of the dressing provided as part of the majority of gauze-based NPWT commercial kits, a PHMB-containing dressing. Initially, the authors thought NPWT could manage infection as the result of removing exudate from the wound; however, they observed bacterial loads increased in contaminated wounds, and found PHMB dressings had no effect on bacterial loads in contaminated wounds (10⁵ CFU/mL) treated with NPWT.⁹ 

When designing this pilot study protocol, the authors took into consideration the available experimental data about the technology for BFBM dressings.^7,8,13-19

Malmsjö et al^7,15 found BFBM dressings transmit negative pressure as efficiently as foam and gauze. Blood flow was found to be similarly decreased by BFBM dressings and gauze but less than with foam. More liquid transferred from the wound when using BFBM dressings and foam, because of the hydrophobic characteristics of the material, and it may explain previous findings that granulation tissue formation is more rapid under BFBM than under gauze.¹⁵ Another advantage is that granulation tissue does not grow into the BFBM dressing, thus the risk of the dressing adhering to the wound may be avoided. 

The BFBM dressing is known to bind and inactivate a wide range of bacteria as well as fungi, and it has been shown to reduce microbial loads without the development of resistance among microorganisms.⁸ The clinical implication of all these findings is that the BFBM dressing could be a better clinical choice than gauze when a mesh-like dressing is required during NPWT. 

The comparison between BFBM dressings and foam was not as straightforward. This is because the clinical response under NPWT to each of these dressings is different. The BFBM dressing seems to be a better choice for shallow noncontracting wounds, complex deep cavity wounds, and for wounds where vascularization and pain are a concern.⁸

It is interesting that BFBM dressings behave differently from conventional antimicrobial dressings. A paradox exists regarding its mechanism of action; microorganisms are trapped, not destroyed, and eliminated from the wound at dressing change. This implies a mind-set change to accept that local antimicrobial activity could be obtained without using traditionally known antimicrobial substances.^{7,8,13-15,17-19} The authors used the BFBM dressing as a WCL and kept the PHMB-containing dressing as a wound filler. 

The authors of this study knew using the PHMB-containing dressing as a wound filler in conjunction with the BFBM dressing as wound interface could be challenged as it may be argued that the PHMB contained in the dressing could have some antibacterial effect. However, it was decided to go ahead for 3 main reasons: first, the authors’ previous research led them to assume its effect on the wound flora of PUs would be negligible; second, the PHMB-containing dressing wouldn’t be in direct contact with the wound bed; and third, during treatment with NWPT, wound exudate is removed continuously and it is probable that any PHMB in contact with the exudate would be equally removed. On the contrary, it is correct to think that if PHMB had been effectively in contact with the wound bed, the bacterial loads would have been lower than the loads actually found.  

Kirkentep-Møller et al²⁰ and Fazli et al²¹ studied the distribution of bacteria in chronic wounds using sophisticated fluorescence and laser microscopy techniques. Their results suggest a significant correlation between the type of bacteria and their spatial organization within the wound, not on the surface of the wound but in its depth. They found S. aureus was in connection with or on the surface of the wound, whereas P. aeruginosa was situated predominantly inside the wound bed. The authors suggest bacteria present within chronic wounds tend to be aggregated in microcolonies imbedded in self-produced biofilm and situated at different depths within the wound. This is clinically relevant as the evaluation of chronic wounds by swab may underestimate the presence of important pathogens because the swab technique detects mainly bacteria situated in the upper region of the wound. Thus, as the authors decided to explore bacterial population by biopsy they didn’t consider it relevant to perform preliminary tests regarding bacterial populations inside the dressing.  

Published evidence about the effect of dressings on bacterial loads is not abundant; evidence considering the effects under NPWT is even more rare. Methodologies are hardly comparable and results variable. The present results contrast with previous scientific evidence about the effect of NPWT on bacterial loads where authors reported either an increase in bacterial load^2-4 or no significant difference when using dressings, with⁹ or without^2,5 antimicrobial properties. Completely opposite evidence has also been presented when considering specific species of bacteria.^2,5 The common point being that NPWT seems to have a positive effect on wound healing. 

This study’s results show a clear bacterial load reduction in the majority of treated wounds in groups B and C. Wounds in group A (with negative or low colonization levels at B0) kept low bacterial numbers across the evaluation. The bacterial load reduction is statistically significant when considering all the patients as a group, but the significance is even more important when considering wounds in groups B and C separately. Conversely, the bacterial loads in the dressings were consistently high at day 1 and remained high at day 7. 

Keeping in mind the differences in the bacterial quantification methods, these results seem to be in accordance with the work performed by Gentili et al,¹³ who in 2012 published the results of a 4-week clinical evaluation in 19 patients (20 wounds) presenting hard-to-heal vascular leg ulcers, treated with BFBM dressings. They used quantitative real-time polymerase chain reaction (PCR) to assess bacterial loads. The results showed treatment with the BFBM dressing promoted healing in 7 patients and provided improvement in another 8. The initial bacterial load was considerably different in the samples ranging from 4.38 x 10³ to 2.44 x 10⁸ bacterial genomes/mg of tissue. Nevertheless, the average of the total bacterial load before the treatment was 4.41 x 10⁷/mg of tissue, which decreased to 1.73 x 10⁵/mg of tissue, corresponding to a significant 254-fold decrease in the total bacterial load in the healing wounds. Whereas in the nonhealing wounds, an insignificant 5.3-fold decrease of the total bacterial load was found. The results allowed them to confirm the suitability of real-time PCR quantification of total bacterial load as a quick and sensitive parameter of wound evolution when performed on tissue biopsies but not on dressing samples. Surprisingly, the analysis of bacterial loads in the dressings in all the patients showed less variability. The majority of samples yielded initially 10⁹ total bacteria and remained similarly high in the final samples. Trying to explain this finding, Gentili et al¹³ suggested it is possible chronic wounds, even at an advanced stage of healing, are still colonized by a high number of bacteria at the surface, while infection of deep tissues has been eliminated.

All this evidence supports tissue biopsy as the best method for quantitative bacterial assessment. However, aiming to provide consistency to this study’s research, the authors decided to evaluate the area of the dressings situated over the region of the wounds where biopsies were taken (ie, the middle of the dressing).

Patients were classified into 3 groups after completing the first set of biopsies for several reasons: first, bacterial loads may correspond clinically to colonization, critical colonization, and local infection; second, the absence of supporting evidence on the subject made it difficult to establish a hypothesis; and third, this evaluation is a pilot study about a not fully defined clinical entity, and the authors needed to establish a baseline for comparison purposes.  

In a 1967 study about skin graft survival in humans, Krizek et al²² showed an average of 94% of grafts survived when ≤ 10⁵ CFU/g were present in tissue biopsies and only 19% survived when the count exceeded 10⁵ CFU/g. Robson et al²³ defined infection as a level of > 10⁵ microorganisms/g of tissue. Using quantitative bacteriology they found wounds undergoing delayed closure with ≤ 10⁵ CFU/g healed successfully, but those with ≥ 10⁵ CFU/g did not. 

In 2001 Gardner et al²⁴ published research about the validity of clinical signs to identify local infection in chronic wounds. Using Robson and coauthors’²³ definition of infection, they assessed 36 chronic wounds and found 11 infected wounds not displaying traditional signs of infection. Gardner et al²⁴ concluded that delayed healing, discoloration of granulation tissue, friable granulation tissue, pocketing at the base of the wound, and wound breakdown were better indicators of chronic wound infection than the classic inflammatory signs. Bill et al²⁵ evaluated the level of bacteria in 38 nonhealing wounds without clinical signs of infection. Tissue biopsies showed 74% of these nonhealing wounds contained > 10⁵ CFU/g of tissue.

The level of bacteria that inhibits wound healing without the standard clinical signs of infection was termed “critical colonization” by Davis in 1996.²⁶ The term colonization has been frequently challenged²⁷ but has not been disproved, and the underlying concept has possibly been used for a long time as part of the wound care lexicon. Some probable synonyms to critical colonization include: silent infection, covert infection, occult infection, refractory wound, subclinical infection, indolent wound, stunned wound, subacute infection, and recalcitrant wound.²⁷ This reflects the importance of recognizing in clinical practice the existence of a condition in chronic wounds that signals a microbiological imbalance. Based on the above, and as more research on how to link clinical signs of infection to quantitative microbiology in chronic wounds becomes available, the authors consider the use of their proposed classification appropriate.

Negative biopsies are not a reflection of technical difficulties experienced during the evaluation but a clear consequence of the pathophysiology of chronic wounds, where microorganisms are not necessarily invading the host’s tissue even if they are present in high numbers at the wound surface. Bacterial loads were ≥ 10⁶ CFU/mL in 75% of the dressing samples processed; these results indicate that vortex spinning the samples in saline solution is effective to some level in extracting bacteria from the dressings. Additionally, results showed a correlation between bacterial species isolated from wound bed biopsies and bacterial species found in the dressings in 77% of cases in group B and in 100% of cases in group C. From these findings, the authors hypothesized that as the bacterial population increased inside wounds and into the host’s tissue, the distribution of bacteria on the surface and on the depth of the wound bed became identical. This observation is supported by the results of Gentili et al,¹³ which also indicates controlling bacterial population on the wound surface may prevent invasion potentially leading to critical colonization or infection. 

The results of this study may indicate that BFBM dressings, by trapping and removing superficial bacteria, help prevent further bacterial multiplication, which may support the host’s defenses to gain control over invading microorganisms.

So far the authors have treated more than 500 categories 3 and 4 PUs with NPWT. They have developed a reliable protocol to measure bacterial load by performing just 1 punch biopsy without anesthetic in the middle of the wound with similar results to those obtained from multiple biopsies and the added benefit of avoiding additional trauma to the patient. This protocol is only used in patients with categories 3 and 4 PUs in the sacral or trochanteric areas, which facilitates performing painless punch biopsies because they are, in general, large wounds with a decreased or absent sensitivity to pain particularly at the center of the wound bed. 

The predominance of gram-negative bacteria in the evaluation could be explained by the fact that 94% of these PUs were located in the pelvic area (86% in the sacrum), which may facilitate fecal contamination of wounds.

Not collecting data regarding wound evolution could be criticized; however, the authors decided to focus on the bacterial response to treatment due to the objective method of evaluating bacterial response to treatment. Also, an observation period of only 7 days does not provide a clear idea about wound evolution when it may take months for a chronic wound to heal.

The safety profile of the BFBM dressing is excellent, proven by the fact that there were no adverse events related to its utilization.

It is imperative that the approach to preventing and managing wound infections be changed. The World Health Organization recently published a report²⁸ about bacterial resistance to antibiotics. Very high rates of resistance (> 50%) have been observed in all regions of the world in common bacteria (eg, E. coli, Klebsiella pneumoniae, and S. aureus). This implies common skin infections may need second-line drugs in many countries. Patients with infections caused by resistant bacteria generally have an increased risk of worse clinical outcomes and death, not to mention consuming more health care resources.²⁸ The possibility of preventing and treating wound infections without the unwanted risk of inducing resistance should be explored. With its interesting properties, the BFBM dressing needs to be considered as part of well-designed strategies and protocols to prevent the development of bacterial resistance.

The BFBM dressing seems to contribute to a statistically significant reduction of bacterial burden in wounds with moderate or high levels of colonization and may have prevented an increase in levels of colonization in wounds with negative or low bacterial loads.

The results of this study combined with available clinical and experimental evidence support the use of BFBM dressings as a contact layer (wound interface) in the management of categories 3 and 4 PUs with NPWT, when the reduction of the wound bioburden is a treatment objective. The BFBM dressing is safe to use in patients with chronic wounds in combination with NPWT. 

Further research to accurately confirm the efficacy of the dressing and assess the evolution of the treated wound would be necessary. It would need to include a longer observation period, the use of BFBM dressing also as a wound filler, the utilization of PCR techniques, and evaluate the response of critically colonized or infected chronic wounds to other types of dressings when used in NPWT. 

Affiliations: Centro Aziendale di Riparazione Tissutale, Wound Care Centre, ASL NAPOLI 3 SUD, Napoli, Italy; and Micobiology Unit, San Leonardo Hospital, ASL NAPOLI 3 SUD, Napoli, Italy

Correspondence:
Marino Ciliberti, MD 
Centro Aziendale di Riparazione Tissutale
Tissue Repair Centre
ASL NAPOLI3SUD via Fusco 12 -80058- Torre Annunziata
Napoli, Italy
marinodoct@gmail.com 

Disclosure: The authors disclose no financial or other conflicts of interest.