Measurement of vancomycin hydrochloride concentration in the exudate from wounds receiving negative pressure wound therapy: a pilot study

Abstract

It has been reported that negative pressure wound therapy (NPWT) is effective in the treatment of contaminated wounds. We hypothesised that systemically administered antibiotics migrate to wound site effectively by NPWT, which provides the antibacterial effect. We measured and compared the concentrations of vancomycin in the exudate and blood serum. Eight patients with skin ulcers or skin defect wounds who were treated with NPWT and were administered an intravenous drip of vancomycin were enrolled in this study. The wound surfaces were muscle, muscle fascia or adipose tissue. We administered vancomycin intravenously to NPWT patients (1–3 g/day). The exudate was obtained using 500 ml V.A.C. ATS ^® canisters without gel. Three days later, the concentrations of vancomycin were measured. The mean concentration of vancomycin in the exudate from NPWT was 67% of the serum vancomycin concentration. We found that concentrations of vancomycin in NPWT exudates are higher than the previously reported concentrations in soft tissue without NPWT. The proactive use of NPWT might be considered in cases of suspected wound contamination when a systemic antibiotic is administered.

Introduction

The use of negative pressure wound therapy (NPWT) for infected wounds is not universally accepted. However, there are reports that NPWT is effective in the treatment of open sternal wounds and open fractures 1. It can be demonstrated that if there is wound contamination, NPWT has an effect on local infection control and therapy. We consider this effect of NPWT to be obtained through several mechanisms. It has been reported that NPWT resulted in a significant decrease in the number of bacteria 2, 3. We hypothesise that antibiotics that are administered systemically migrate to the wound site effectively by NPWT, resulting a reduction in the number of bacteria. Pharmacokinetically, it is thought that the local concentration of antibiotics works directly to create the antibacterial effect.

To measure the concentration of antibiotics that perfuse to the wound during NPWT, we focused on the exudate that is collected in the canister of the NPWT system. The concentration of vancomycin hydrochloride in the exudate is theoretically equivalent to that in the interstitial fluid surrounding the wound tissue. In addition, we measured the concentration of serum vancomycin (which is frequently measured in clinical practice).

Frequently, methicillin‐resistant Staphylococcus aureus (MRSA) infection of the wound becomes complicated. Healing takes a long time. Vancomycin is the drug of first choice for MRSA. However, it is well known that it has a low penetrance to skin and fat tissue and its effect on wound colonisation and infection is sometimes limited.

We measured and compared the concentrations of vancomycin in the exudates and blood serum to assess the effect of NPWT on the migration of vancomycin from the blood stream to the wound site by comparing its value with previously reported values. Our hypothesis is that there is some effect of NPWT on the vancomycin concentration of the exudates.

Patients

Eight patients with skin ulcers or skin defect wounds (five men and three women; aged 24–83 years) who were treated with NPWT were enrolled in this study after approval of our institutional review board. The wound surfaces were muscle, muscle fascia or adipose tissue. There were no clinical signs of significant infection in any of the wounds.

There was one patient with diabetes and one with hypertension. There were two lower leg ulcer wounds (the wound surface was adipose tissue), two skin/subdermal tumour‐removal wounds (adipose tissue), two fasciotomy wounds secondary to compartment syndrome surgery (muscle and muscle fascia), one third‐degree burn (adipose tissue) and one post‐debridement surgical site infection (adipose tissue). For this last patient, the wound was previously opened and drained and the infection controlled.

Methods

First, we used the Cockcroft–Gault formula‐derived creatinine clearance to determine the dosage of vancomycin, which was then administered intravenously to NPWT patients.

For NPWT, we used the V.A.C. ATS^® system (KCI KK, Tokyo, Japan). The exudates were obtained using the 500 ml V.A.C. ATS^® canisters without gel. The canisters had a hole at the top that vacuumed the air; a needle was placed through this hole for sampling the exudates. Three layers of protective film were applied when using the V.A.C. ATS^® system to prevent water vapour condensation resulting from air leakage.

After the third day, the concentrations of vancomycin were measured. The vancomycin concentration of the exudates was measured along with the trough level prior to vancomycin administration. The fluorescence polarisation immunoassay method was used to measure the concentration of vancomycin. After the exudates were sampled, impurities were precipitated using a centrifuge. The supernatant was measured using the above process.

The verified stability of vancomycin was high. Vancomycin was dissolved in several solutions and stored at 24°C for 24 hours; the final titre of vancomycin was 97–100% of the initial titre. There was essentially no degradation of vancomycin 4.

The pharmacokinetic parameters of vancomycin were calculated by VCM‐TDM Microsoft‐Excel Version2.03 (Practical Pharmacological Program, designed by Shionogi & Co., Ltd., Osaka, Japan).

Area under the blood concentration time curve (AUC) was determined using the following equation for AUC 5:

$$\mathrm{AUC}\left(\mathrm{mg}•\mathrm{h}/\mathrm{l}\right)=\mathrm{Dose}\left(\mathrm{mg}\right)/{\mathrm{CL}}_{\mathrm{vanco}}\left(\mathrm{l}/\mathrm{h}\right)$$

We calculated the average serum vancomycin concentration from the AUC as follows:

$$\begin{array}{cc}& \mathrm{The}\mathrm{average}\mathrm{serum}\mathrm{vancomycin}\mathrm{concentration}\left(\mathrm{\mu g}/\mathrm{ml}\right)\\ & =\mathrm{AUC}/\mathrm{h}=\left\{\mathrm{Dose}\left(\mathrm{mg}\right)/{\mathrm{CL}}_{\mathrm{vanco}}\left(\mathrm{l}/\mathrm{h}\right.\right./\mathrm{h}\times 1000\end{array}$$

Haematocrit and haemoglobin levels in the exudates were measured. The haematocrit was 2–5% and haemoglobin was 0·8–1·5 g/dl. We calculated the theoretical concentration of vancomycin in exudate using the following corrective formula 6:

$$\begin{array}{cc}& {\mathrm{C}}_{\mathrm{E}}=\left\{\left(100-{\mathrm{H}}_{\mathrm{D}}\right){\mathrm{H}}_{\mathrm{B}}{\mathrm{C}}_{\mathrm{D}}-\left(100-{\mathrm{H}}_{\mathrm{B}}\right){\mathrm{H}}_{\mathrm{D}}{\mathrm{C}}_{\mathrm{B}}\right\}/\\ & 100\left({\mathrm{H}}_{\mathrm{B}}-{\mathrm{H}}_{\mathrm{D}}\right)\end{array}$$

where C_E is the theoretical concentration of vancomycin in exudate, C_D the actual concentration of vancomycin in exudate, C_B the concentration of vancomycin in serum, H_D hematocrit in exudate and H_B hematocrit in blood. There was no effect of mixing with blood.

Results

The detailed patient information is provided in Table 1. The mean age was 44·3 years old. The mean daily dosage of vancomycin was 2375 mg. The serum trough concentrations of vancomycin, the concentrations of vancomycin in the exudates and vancomycin clearance values (CL_vanco) are shown in Table 2.

Table 1.

Patient background and daily vancomycin dosage

		Age	Sex	Location	Daily vancomycin dosage (mg)
1	Skin ulcer	72	F	Lower leg	500 × 2
2	Skin ulcer	41	M	Lower leg	1500 × 2
3	Subdermal tumour removal	49	M	Hip	1500 × 2
4	Skin tumour removal	28	F	Neck	1500 × 2
5	Post‐fasciotomy due to compartment syndrome	24	M	Lower leg	1000–1500 × 2
6	Post‐fasciotomy due to compartment syndrome	24	M	Lower leg	1500 × 2
7	Third‐degree burn	68	M	Arm	500 × 2
8	Surgical site infection	48	F	Chest wall	1250 × 2
	Mean	44·3			2375

Table 2.

Serum and exudate vancomycin concentrations and percentages

	Trough serum vancomycin concentration (µg/ml)	Exudate vancomycin concentration (µg/ml)	Vancomycin clearance(l/hour)	Average serum vancomycin concentration (µg/ml)a	Exudate vancomycin/average serum vancomycin (%)
1	12·28	14·42	1·168	27·73	52
2	17·6	14·1	4·852	27·28	52
3	12·46	16·11	5·216	25	64
4	73·2	71·4	0·954	115·86	62
5	9·5	13	6·539	14·94	87
6	6·5	17·4	7·469	16·74	104
7	20·1	13·5	2·338	26·73	51
8	11·1	14	4·797	21·71	64
					Mean 67%

^aWe calculated the average serum vancomycin concentration using the formula: The average serum vancomycin concentration (μg/ml) = {Dose (mg)/CL_vanco(l/h)}/h × 1000.

Table 2 also shows the percentages of vancomycin concentration that perfused to the exudate and the average serum vancomycin concentrations. The mean percentage of perfusion was 67%.

Discussion

We investigated the effects of NPWT on the concentration of vancomycin hydrochloride in wound exudate during the therapy. Results showed that the mean vancomycin concentration in exudate from NPWT is 67% of the serum vancomycin concentration. This is higher than the previously reported migration rate of vancomycin into the interstitial fluid without NPWT 7. These reported concentrations of vancomycin in the interstitial fluid were calculated by the microdialysis method. A flexible microdialysis probe with a membrane was inserted under the interstitium of soft tissue and perfused with Ringer's solution by means of a microinfusion pump. Microdialysis samples for determination of vancomycin concentration were collected and measured. It is thought that the measurement of the local tissue concentration of vancomycin by this method is very accurate. This report 7 concluded that the vancomycin interstitial fluid concentration was 10% of the serum concentration in diabetics and 30% of the serum concentration in non‐diabetics.

NPWT has been used in various clinical situations, including trauma such as soft tissue injuries, open fractures and burns; it is also effective for surgical procedures such as cutaneous fixation of split‐thickness skin grafts 1.

There have been various reports regarding the mechanism of the effects exerted on the wound 8, 9. These mechanisms include eliminating excess exudate 10, promoting granulation tissue proliferation 11, decreasing the bacterial count 2, 3, contraction of the wound edge 10 and maintaining a moist environment 10.

We would like to discuss two effects that are related to the mechanisms involved in this investigation. First, elimination of excess exudate is achieved and cutaneous blood flow is also promoted. The exudate is removed from the surface of the wound, and due to the negative pressure gradient applied to the interstitial fluid, interstitial oedema is reduced and the blood flow surrounding the wound is improved. We have found no reports regarding the quantitative improvement in interstitial oedema. Wackenfors et al. reported that blood flow to muscle tissue 2–3 cm away and the skin tissue 2·5–3·5 cm away from the wound margin was increased by 1·5‐fold to 2‐fold 12.

Next, the effects that reduce the bacterial count are present. NPWT has also been demonstrated to be effective against bacteria at the wound surface. A reduction in the bacterial count was observed in the wound tissue culture and results showed that the effects against gram‐negative bacilli and Pseudomonas aeruginosa were greater than for S. aureus 2, 3. Clinically, NPWT was used to treat vascular graft infection, and investigators described swabs from the wound surface as being negative 13.

NPWT is indicated for a wide variety of diseases and injuries, but there are also instances where it is contraindicated. Contraindications include cases where necrotic tissue remains in the wound surface, cases where there is haemorrhage, massive infections and cases where there is a fistula to visceral organs 9. There are reports of toxic shock syndrome onset when NPWT was used to treat infected wounds 14, so care should be taken when using NPWT where there is already contamination.

However, there are some cases in which NPWT is recommended once the infection has subsided to a certain extent. In particular, NPWT was effective in the case of mediastinitis or sternal osteomyelitis after debridement was performed and the infection was controlled 1. Furthermore, Lui et al. reported that in cases of Gustilo IIIb open fractures, the rate of infection was significantly reduced within 7 days of NPWT, when bone fixation and debridement were conducted immediately 15. Based on these reports, NPWT is considered useful in preventing infection even if there is bacterial contamination, but not when critical colonisation occurs.

We then considered whether NPWT resulted in more than a mere decrease in the bacterial count and whether systemic administration of antibiotics via the NPWT device contributes to local effects on the wound. In terms of pharmacokinetics, systemically administered antibiotics migrate to the tissues and exercise their effects. Antibiotic effects are dependent on the drug concentration and it is believed that the effects increase when the local concentration increases. We considered whether the amount of antibiotic that migrates to the wound site is increased through NPWT.

Exudate that is obtained from NPWT has approximately the same electrolyte concentrations as that of interstitial fluid 16; thus, exudate and interstitial fluid can be said to be equivalent theoretically. In other words, we can conclude that if the antibiotic concentration in the exudate rises, then the interstitial fluid concentration increases. We decided to compare the antibiotic concentration in the exudate and the plasma concentration in this study.

Furthermore, we measured the concentration of vancomycin, an antibiotic that is commonly used in clinical practice. Vancomycin is a glycopeptide antibiotic and is effective against MRSA infection and critical colonisation, which are becoming clinically problematic. Vancomycin was also selected because the fluorescent antibody assay method can be used with this antibiotic. Vancomycin is known to have a low migration rate into some tissues, with reported migration rates of 86–88% into pleural fluid 17, 41% into the lungs 18, 23·8% into bone tissue and 50·8% into bone marrow blood 19. Migration rates into adipose tissue and interstitial fluid are 21·6% 20 and 30% 7, respectively, with a low migration rate into cutaneous tissue. When these reports of migration rates are compared to our results, the vancomycin concentrations in NPWT exudates are high; this indicates that the migration rate is increased by NPWT.

To the best of our knowledge, the reports that investigated the antibiotic concentration in the exudate of human skin defects and surgical wounds involved penicillins 21, 22, cephalosporins 23, 24, 25 and carbapenems 26, and in terms of anti‐MRSA drugs, the concentration of linezolid was only measured in blisters 27. We did not find any reports that investigated the migration of vancomycin into wound exudates. Apparently, there have been no reports that have measured the antibiotic concentration in exudates from NPWT except for this report.

The mechanisms by which NPWT affects the rate of antibiotic migration can be inferred from the following two points. The first is the effect of filtration under reduced pressure. Normally, substances with a molecular weight of less than 5000 are able to move freely through the capillaries. Vancomycin (molecular weight: 1485·71), which is not bound to plasma proteins, is thought to migrate into the interstitial fluid via passive diffusion along its concentration gradient. The vancomycin protein binding rate is said to be 34% 28 and approximately half of the unbound vancomycin migrates into the interstitium. Negative pressure applied to the interstitial surface results in filtration at reduced pressure. The interstitial fluid is consequently drawn from within the blood vessels, and the low‐molecular‐weight substances are drawn from the capillaries. Filtration itself is not specific to vancomycin but may increase its migration speed.

The second effect of NPWT is the previously mentioned action of increasing perfusion in the surrounding skin. Because of increased perfusion, the amount of vancomycin that is transported to the capillaries increases and this may increase the total amount of vancomycin that migrates into the interstitial tissues. When we investigated the differences in the wound surfaces in this study, some small differences in the migration rate occurred depending on whether there was muscle, muscle fascia or adipose tissue at the wound surface (Figure 1). This suggests that the differences in perfusion between the original tissues may be reflected in the results.

Figure 1.

Vancomycin concentrations in wound exudates are higher than the previously reported migration rate of vancomycin into the interstitial fluid. : wound surface is muscle or muscle fascia. , wound surface is adipose tissue. , previously reported vancomycin migration rate into the interstitial fluid.

Based on the above results, NPWT and systemic administration of antibiotics, at least regarding vancomycin, might be effective in the treatment of wound contamination. However, limitations of this study included the fact that sample size was small and that changes in the migration rate may have occurred according to the wound type and the use of different antibiotics. Further study using a larger sample of cases is required.

In conclusion, we found that vancomycin concentrations in wound exudates from NPWT are higher than the previously reported concentrations without NPWT. The proactive use of NPWT might be considered in cases of suspected wound contamination when a systemic antibiotic is administered.

Ida Y, Matsumura H, Onishi M, Ono S, Imai R, Watanabe K. Measurement of vancomycin hydrochloride concentration in the exudate from wounds receiving negative pressure wound therapy: a pilot study.