Multiple Mechanisms of NPWT

Contrary to what you might believe it is vital for wound care companies to have a completely unbiased view of the published scientific literature relating to their industry. Here I share my recent thinking as we enter a period of quickening development into the science behind the clinical effects of NPWT (Negative Pressure wound Therapy). It doesn’t pretend to be a comprehensive review but sets out the critical components for understanding the mechanisms.

NPWT might justifiably be said to have entered the mainstream of the clinical and scientific community with the publication of back-to-back papers in the same issue of the peer reviewed journal Annals of Plastic Surgery in 1997.1,2 In the first paper Morykwas et al describe experiments in pigs designed to explore and demonstrate the scientific mode of action of NPWT using open cell polyurethane (PU) foam as the filler which many will know simply as “black foam”. In the second paper Argenta et al describe a case series of some 300 patients treated with NPWT using open cell PU foam on a range of acute, sub-acute and chronic wounds. Since much of the literature on the scientific understanding of the mechanism of action (MOA) stems from the first Morykwas paper, it is appropriate to review that paper in a little detail.

The Morykwas paper reported four experiments. In the first experiment full thickness wounds were created on the backs of pigs. A laser Doppler probe was inserted into the healthy peri-wound skin to record any changes in blood flow. NPWT using PU foam was applied to the wounds at a range of pressures (-50, -75, -100 mm Hg etc) and the effects on blood flow were recorded. Increases in the Doppler signal were recorded. The effects last for 5-10 minutes. The greatest response was seen at -125 mm Hg. This pressure was then selected for further experiments. It is important to realize this is the only published evidence for recommending -125 mm Hg as the optimal pressure for NPWT.

In the second experiment full thickness wounds were again created on the backs of pigs; some wounds were connected to NPWT at -125 mm Hg and some were left as controls. Each day the animals were anesthetized and the volume remaining in the wound was measured by filling it with dental impression material, which was allowed to set and then removed to measure its volume. Note that on the backs of these pigs there was very little sideways wound contraction and these wounds filled entirely by granulation from the base. There was a statistically significant increase in granulation compared to the control wounds of some 60% after 10 days. (There were also some experiments where NPWT was applied intermittently and the granulation tissue grew even faster in those wounds. We will not review the implications for MOA and intermittent therapy in the interests of space, but this certainly a subject for future investigation.)

In the third experiment full thickness wounds were deliberately infected with bacteria (Staphylococcus aureus) so that the level of bacteria per g of tissue was above 105 (the threshold above which clinically wounds will not heal without some specific intervention). NPWT was applied to some wounds and others were left as controls. The level of bacteria in the wounds treated with NPWT was reduced more than 100 fold from over 105 to 103 per gram tissue.

The fourth experiment is rarely discussed but is perhaps the most remarkable of them all. On the flanks of anesthetized pigs rectangular random pattern skin flaps were created were the blood supply to the flap came from the single remaining upper edge. The flaps were deliberately created so that there was insufficient blood flow from the intact (proximal) edge and over 4-5 days the flaps suffered necrosis from the distal end to about 50% of their length. When NPWT was applied to the flaps after they had been created, they survived to about 65% of their length. Remarkably, when NPWT was applied to the intact skin for a 4-day period before the flaps were created, the flaps also survived to about 65% of their length when the flaps were raised from the previously treated skin. When NPWT was used both before and after creating the flaps, the effects were additive and the flaps survived to more than 0% of their length. This fascinating experiment is not well explained by our current theories although some kind of ischemic pre-conditioning mechanism would be a likely candidate.

They say that there is “nothing new under the sun” and this is particularly true for NPWT. Some eight years before the Morykwas and Argenta papers Mark Chariker a plastic surgeon working with nurse Katherine Jeter published an article describing how they used vacuum (at -60 to -80 mm Hg) applied to wounds filled with gauze under film dressings to manage entero-cutaneous fistula wounds.3 Their article described how this NPWT technique was able to facilitate healing and closure. The journal in which the Chariker-Jeter technique was published was a not prestigious peer reviewed journal like Annals of Plastic Surgery used by Morykwas & Argenta (you won’t find Chariker-Jeter it in a PubMed search) and although the Chariker-Jeter technique was used locally it was largely unnoticed until more recent times.

So how do foam and gauze operate under NPWT? Initially porcine model experiments at the University of Lund, Sweden showed a 1:1 relationship between the negative pressure applied and the negative pressure transmitted to the wound bed whether the wound is filled with foam or gauze.4 When the clinical effectiveness of NPWT applied with gauze at -80 mm Hg was reviewed in retrospective data from 30 patients wounds reduced on average by 15% per week. This is similar to the published rates of wound volume reduction from NPWT applied using foam at -125 mm Hg.5

So it seems that there is data showing both -125 and -80 mm Hg are clinically effective with different filler materials. What evidence is there for an optimum pressure? The answer is virtually none, but one paper that will often be cited is a study published in 2001.6 In this study pigs were treated with foam based NPWT at -25 mm Hg, -500 mm Hg, neither of which stimulated granulation and -125 mm Hg, which did. So this study does show that very low and very high negative pressures are not optimal for NPWT but the study says nothing on the relationship around the range -50 to -150 mmHg within which clinicians might operate.

Is there any clinical evidence for an optimal negative pressure for effective NPWT? There are no direct studies but in a 2007 paper the clinicians did record what happened when they used different levels of negative pressure in different patients.7 McCord et al were reporting their excellent experiences using foam based NPWT on paediatric ICU patients who had pressure ulcers, reconstructive surgery or dehisced surgical wounds. In order to minimize pain in the children, the clinicians used their judgment and selected lower pressures if their patients experienced pain. Their results showed no relationship between pressure applied and NPWT at -50, -75, -100 or -125 mm Hg.

To summarize at this point both foam and gauze appear to be clinically effective in delivering NPWT. Pressure is an important variable but its relationship to outcome is not clear and the evidence suggests it is effective in the range -50 mm Hg to -150 mm Hg. What progress has been made since Morykwas & Argenta’s papers in 1997 towards understanding the way in which NPWT achieves it clinical effects?
The published evidence around NPWT can be grouped around five areas. (Note this doesn’t necessarily mean that all of these are legitimate mechanisms of action.)

Stretching and holding;
Effects at the wound interface;
Effects on blood flow;
Effects on EDEMA;
Effects on bacteria.

Stretching and holding
NPWT is a primarily a physical effect. Stretching and holding forces play a role and these are often termed macro-mechanical effects. Before we consider some of the evidence on the stretching and holding of tissue by NPWT it is worth taking the time to distinguish between two phrases which are used in the literature: micro-deformation and macro-deformation. Micro-deformation means the transfer of an imprint of the surface topography of the compressed wound filler to the tissue surface. More on this shortly. In contrast, macro-deformation means the stretch or contraction of tissue when NPWT is applied. Some have used the phrase “reverse tissue expansion” to describe NPWT in analogy to the use of tissue expansion in plastic surgery where balloons are placed beneath the skin and gradually filled with air or fluid to stretch the skin. Under the stretch force cells undergo increase proliferation and production of matrix and increased skin tissue results.
In a 2008 paper from a group in Japan, direct video microscopy was used to observe what happed to blood vessels in exposed skin in anesthetized rats. When NPWT was applied such that the field containing the blood vessels was stretched, the lumen of the vessels was seen to increase and this increase was accompanied by increased blood flow.8

Another important way in which the physical force of holding tissue with NPWT is used to great effect is in the application of split thickness skin grafts. A great example is a study from Chile, which was conducted without any commercial involvement as a randomized controlled trial of NPWT versus control bolster dressings.9 The group actually used wall suction rather than portable pumps and they used their own filler/interface which consisted of a gauze and closed cell foam dressing. The researchers recorded the area in cm2 of graft that did not take and there was a clear difference in the amount of graft lost in the control group compared to the negative pressure closure group. This effect is almost certainly a purely physical action, which protects against the two main enemies of skin graft take: sheer between the graft and the bed and the accumulation of fluid beneath the graft. Both interrupt the growth of new blood vessels from the wound bed into the graft and as NPWT is able to eliminate movement between the graft and the accumulation of fluid, skin graft take is significantly improved.

Effects at the wound interface
In contrast to macromechanical effects like stretching and holding, micromechanical effects take place right at the interface between the filler and the wound bed.
One of the most common observations in the use of NPWT is the surface stimulation of granulation tissue beneath open cell polyurethane foam. This phenomenon has rightly attracted researchers as it is a relatively rapid response, which demands an explanation. The greatest contribution to the phenomenon has come from the laboratory in Boston headed by Plastic Surgeon Dennis Orgill. In the 2004 paper from this group Saxena et al modeled the forces which might be applied to wound tissue by an individual pore in open cell foam when it was compressed under negative pressure.10 Using values for the stiffness of foam and tissue the group constructed a 2D computer model that predicted how the wound surface might be micro-deformed by foam at different negative pressures and their results are a good match for what is seen in practice.

So is micro-deformation a phenomenon that is exclusive to open cell foam? Probably not. In a paper published recently, Wilkes et al describe a more sophisticated 3D computer model of the way that NPWT filler (foam or gauze) deforms the tissue surface during the application of NPWT.11 A special technique called micro-CT was used to image the interface between the foam or gauze with an artificial tissue and the computer model was used to calculate the micro-deformational forces. Gauze seems to deliver more than 80% of the micro deformation that foam produces. Thus both foam and gauze seem to act through a similar mechanism and it is not surprising to see foam gives more surface deformation given its property to show more rapid surface granulation effects on tissue than gauze.

Effects on blood flow
At start of this article I reviewed the original experiments published by Morykwas et al in 1997.1 The first experiment looked at the stimulation of blood flow in pigs. What has been learned about blood flow since then? In 2004 a paper was published from a group at the University of Lund in Sweden that used similar techniques to that used by Morykwas several years before. Laser Doppler probes were placed in the peri-wound tissue adjacent to wounds in pigs.12 The new study showed that when the Laser Doppler probe was inserted close to the wound edge blood flow was not increased when negative pressure was applied, on the contrary, it decreased. It was not until the probes were more than 2 cm away from the edge that blood flow increased when negative pressure was delivered. Further work has confirmed this pattern is the same for both foam and gauze filled wounds.13 Of course we don’t know whether the clinical effects would be maximized if the hypo-perfused zone was greater or smaller. However, in biology generally when new blood vessels grow (the process of angiogenesis) they grow from a region of high tissue oxygen into an area of hypo-perfusion or low tissue oxygen. It is tempting to think this might be a mechanism whereby NPWT stimulates angiogenesis in the wound bed.
What mechanism might explain the reduction in blood flow close to the wound bed during NPWT? A paper published last year from South Africa may offer the answer.14 In this study a pressure monitor was placed in the wound bed (about 0.5 to1.0 cm deep) in patients receiving foam based NPWT at -125 mm Hg. So do we see negative pressure in the tissue? We do not. On average Kairinos et al see a small positive pressure (around +5 to +10 mmHg), which seems to reduce to zero over 2 days. Could this small positive pressure compress capillaries near the wound edge and give rise to the pattern of relative hypo-perfusion that is observed in the Laser Doppler experiments from Sweden?

So is there any negative pressure transmitted from the wound space to the tissue during NPWT? Again in a new paper published in 2009 Murphey et al used a micro-manipulator to lower a thin interstitial fluid pressure probe through a film dressing and foam delivering NPWT to a exposed rabbit muscle.15 At -125 mm Hg as the probe enters the tissue, negative pressure is detected, but as the probe is lowered deeper the pressure quickly reduces so that by the time the probe is 0.250 mm (250μmm) deep the pressure is only -60 mmHg. The greater the pressure level, the deeper the negative pressure zone extends, but in all cases the depth at which any negative pressure penetrated was less than 1mm of tissue.

Effects on EDEMA
Tissue EDEMA is produced upon traumatic injury, burns or as the result of inflammation, especially in chronic wounds. If NPWT can reduce EDEMA then recovery from injury and inflammation will be quicker and improved oxygen supply will result. Tissue oedema has probably not received the attention it deserves with respect to NPWT. We have performed some pilot experiments (unpublished) with Dr Steve Young and Sylvie Hampton in the UK using high frequency ultrasound to measure non-invasively the level of fluid in the peri-wound tissue adjacent to pressure ulcers. When gauze based NPWT was applied at -80 mm Hg the level of EDEMA was reduced after 2 days and close to normal after 2 weeks. High frequency ultrasound appears to be a promising technique for monitoring NPWT.

Effects on bacteria
Whereas the early animal work by Morykwas et al1 showed convincing reductions in the levels of bacteria in deliberately contaminated wounds, most clinical studies that have looked at the levels of bacteria during NPWT have not seen this effect. For example Weed et al16 and Moues et al17 found increases or steady bacterial loads respectively in wounds that were nonetheless responding very well to foam based NPWT at -125 mmHg. It looks like reduction in bacteria per see it not a major mechanism of action for NPWT.

Conclusions
It seems clear from the published literature that there are multiple mechanisms of action that make up the overall clinical effect of NPWT. Where investigations have taken place it seems that the fundamental mechanisms of action of foam and gauze are the same. Going forward perhaps one of the next phases of the NPWT story will be to investigate how the nature of the interface between NPWT and the wound bed presented by different fillers can be exploited to achieve different clinical goals in different clinical situations.

Dr Robin Martin PhD, is Clinical Science Program Manager, Advanced Wound Devices Strategic Business Unit, Smith & Nephew Medical Ltd. He can be reached at robin.martin@smith-nephew.com