Abstract: The purpose of this clinical study was to investigate the effects of negative pressure wound therapy (NPWT) on the healing of diabetic foot ulcers and to compare this dressing with traditional moist gauze dressing as a treatment used prior to other wound closure techniques, such as flaps or grafts. Twenty-four diabetic patients were randomly divided into two groups: NPWT group and control group. Initially, the mean surface area of the diabetic wounds was 109cm² in the NPWT group and 94.8cm² in the control group. The mean duration of wound care (until the wounds were covered with granulation tissue) was 11.25 days in the NPWT group and 15.75 in the control group (p=0.05); following NPWT or moist guaze dressing, the mean surface area of the wounds was 88.6cm² in NPWT group and 85.3cm² in control group (p<0.05). In conclusion, the use of NPWT may be an alternative therapy to achieve a faster granulating wound bed in diabetic foot ulcers in order to prepare the wound bed for other closure techniques. Further studies are needed to clarify effects and indications and to modify the technique of this treatment for nonhealing wounds.

R

econstruction of diabetic foot ulcers is often a challenging problem. The impairment of the healing process and the lack of resistance against infection in patients with diabetes represent a familiar clinical problem¹. High treatment costs and unsatisfactory results are common.
       The surgical treatment of the diabetic wounds with loss of soft tissue usually consists of closure using split-thickness skin grafts or transposition flaps. However, immediate surgical closure often fails because the general conditions of patient and wound may not be appropriate for surgical closure; therefore, initial step consists of standard wound care (moist gauze dressing) to prepare the wound bed for final closure.
       Negative pressure wound therapy (NPWT) was developed by Argenta and Morkywas to promote healing of open wounds. In clinical and experimental studies, effects of the negative pressure are reported as an increase of local blood flow, formation of granulation tissue, and decrease of bacterial colonization^2,3. The faster wound healing results in decreased hospitalization and avoidance of additional morbidity of chronic wounds^4–6. Successful results of NPWT were also reported in other studies^7–9.
       The aim of this clinical study was to compare NPWT to conventional moist gauze dressings on diabetic foot wounds as a method to prepare the wound bed for final closure techniques.

Materials and Methods
       Twenty-four diabetic patients with nonhealing wounds of the lower extremity were included in the study and randomized prospectively to NPWT, which was a standard medical aspirator system manufactured by Bicakcilar Inc. (Istanbul), or control (saline-moistened gauze dressing). To avoid bias, all patients were prospectively randomized into groups according to the last digit of the hospital protocol numbers, which were given by a blinded official. Patients with odd numbers were assigned to the NPWT group, and patients with even numbers were assigned to the control group. All of the patients provided informed consent, and the study was approved by the institutional ethics committee.
       Seventeen of 24 patients (71%) had insulin-dependent diabetes and seven patients (29%) had noninsulin-dependent diabetes. Fifteen patients (62.5%) had peripheral neuropathy with changing intensities. Five patients (21%) had peripheral vascular dysfunction and underwent revascularization for this reason. One patient was having treatment for chronic renal failure.
       The diabetic foot ulcers were surgically debrided prior to initiation of NPWT or moist gauze dressing. In the control group, traditional moist gauze dressing was used and the dressing was changed twice a day. After debridement, the surface area of diabetic ulcer wound was measured at operating room and bedside by using a sterilized milimetric paper with a disposable graphic pen. Paper was cut to fit the wound. This template was useful for measuring the surface area of wound and for cutting the sponges of NPWT for optimum wound
coverage.
       In the NPWT group, the wounds were covered with polyurethane ether sponge, and a tube was placed underneath the sponge. Tube and sponge were covered with an adhesive, clean, nonsterile drape for obtaining an airtight seal. Then the tube was connected to a medical aspirator pump in order to create a negative pressure. The aspirator pump has an indicator to measure the negative pressure from 50mmHg to 300mmHg and a disposable canister for collecting the aspirated fluid. In this group, 125mmHg of continuous negative pressure was used, and the sponge and tube were changed every 48 hours in correlation with other studies^2,3,7,8,10.
       Before changing sponges, intravenous analgesic agents were given to all patients of the NPWT group because the polyurethane sponges were sticking to the wounds, which caused pain upon removal. In the control group, the dressings were changed twice daily with less pain. Oral analgesics were administered to the control group when required. Systemic antibiotics were given to all patients after surgical debridement for prophylaxis. There were no signs of wound infection requiring antibiotics during the rest of treatment. Multivitamins and nutritional supplements, including zinc, were administered to all patients. The diabetic wound surface areas were measured every 48 hours in both of the groups. Therapy was continued in both groups until the wound beds approached nearly total coverage with granulation tissue without any of inflammatory signs. In this study, wound closure was to be achieved by lesser surgical procedures.
       During the healing process, the patients ambulated using walking sticks and/or wheelchairs to avoid putting pressure on their wounds. While in bed, clinicians wrapped the feet with five-layer cotton bandages and used pillows to elevate the feet. Once the wounds were healed, patients were advised to use special orthopedic shoes.
       All data were collected prospectively. Treatment groups were compared via the Wilcoxon Signed Ranks and Mann-Whitney tests for homogeneity and data for significance with SPSS 10.0 for Microsoft Windows statistical software. Significance was determined at p<0.05 for all tests.

Results
       The mean age of the patients in the NPWT group was 66.2 (54–77 range) years and it was 64.7 (56–74 range) years in the control group (p=0.506). Before the treatment, the mean diabetic wound surface area was 109cm² in the NPWT group and 94.8cm² in control group (p=0.729). There was no significant difference in groups regarding the initial wound surface areas and ages (p>0.05).
       During the first week of NPWT, an increase in formation of granulation tissue and a decrease in nonviable tissue were seen. The edema in the extremities diminished in all patients, and the surface area of wounds decreased. The mean length of treatment was 11.25 days in the NPWT group and 15.75 in control group (p=0.05).
       After the therapy, the mean diabetic wound surface area decreased from 109cm² to 88.6cm² (20.4cm²) in the NPWT group and decreased from 94.8cm² to 85.3cm² (9.5cm²) in the control group (p=0.032). There was a significant difference in decrease rates. NPWT reduced the wound surface areas more effectively than moist gauze dressing (p<0.05).
       Following the NPWT or control therapy, 19 of 24 wounds underwent split-thickness skin grafting, and 4 of 24 required transposition of regional fasciocutaneous flaps for wound closure. One patient in the NPWT group required wound closure by using a distal pedicled fasciocutaneos flap based on sural vein and its concomitant arterial vessels. In the control group, three patients underwent wound closure by using regional distal pedicled fasciocutaneous flaps as two sural flaps and one safenous flap. The flap donor sites were covered using split-thickness skin grafts. All of the operations were carried out under regional (spinal) anesthesia.
       After NPWT in one patient, the wound bed granulated rapidly and was too narrow for surgical closure; therefore, it was allowed to close by secondary intention. All of skin grafts covered the wounds in a satisfactory manner, and there were no complications with the transpositioned fasciocutaneous flaps except one in the control group, which had partial necrosis distal; however, it was repaired by local debridement and split-thickness skin grafting. All data of the NPWT and control groups, including wound records and closure techniques, are listed in the Table 1 and Table 2.
       There were no local or systemic signs of infection in any of the wounds. No negative impact was seen on extremity functions and psychology of patients. Bleeding was experienced by some of the NPWT group patients during dressing changes due to excessive growth of granulation tissue into the sponge; however, applied light pressure was enough to stop bleeding.

Discussion
       It has been reported that NPWT decreases bacterial colonization and interstitial edema and increases capillary blood flow^2,3. The localized negative pressure removes fluid from wound and promotes the formation of granulation tissue, which is required for wound closure^2,3. Furthermore, it reduces wound surface area by the traction force of negative pressure, which increases mitosis of tissue around the wound^11,12.
       Moist wound environment was reported to heal wounds faster by increasing epithelial migration and by decreasing the infection rates. The negative pressure provides a moist wound bed, which is required for faster healing of chronic open wounds^2,3,11,12. Faster wound healing during the treatment of diabetic foot ulcers may decrease hospital stay and prevent extensive surgery for wound closure, which in turn may lower the cost of treatment and additional morbidity of infection and pain⁴.
       Wound debridement is essential before using NPWT to promote healing of diabetic wounds because nonviable tissue has been shown to delay healing^2,3. In our study, all of diabetic ulcers underwent debridement, and wound surface areas were measured after debridement. Before the treatment, the mean diabetic wound surface area was 109cm² in the NPWT group and 94.8cm² in control group (p=0.729). There was a difference in groups regarding the initial wound surface areas and ages, but it was not statistically significant (p>0.05). The randomization of a small number of patients may cause this non-significant difference. However, a significant difference was observed in decrease rates: NPWT reduced the wound surface areas more effectively than moist gauze dressings (p<0.05). These results correlate with other reports^9,10.
       Clare closed six of 17 diabetic and dysvascular wound patients via changing wound dressings and use of growth factors with the NPWT¹⁰. In the study of De Franzo, 71 of 75 lower-extremity wounds with exposed bone were healed by using NPWT, and 12 of the healed wounds were treated by delayed primary closure⁷. De Lange studied 100 patients with multiple regional wounds and treated 29 of them by using NPWT without surgical intervention⁸. In our study, no growth factors were used but multivitamins and nutritional supplements, including zinc, were given to all of the patients.
       In the experiences of Josty on degloving injuries, the length of NPWT was 4 to 8 days¹³. Clare applied NPWT for an average length of 8.2 weeks (1 to 20 weeks).10 The mean duration of our study was 11.25 days in the NPWT group and 15.75 in the control group (p=0.05). The wounds in the NPWT group were prepared for surgical closure significantly faster than the control group. The duration of therapy in our study was shorter because NPWT was ended when the wound bed approached nearly total coverage with granulation tissue without any inflammantory signs^4,10. In correlation with Armstrong, NPWT was used for the smallest duration as a prior therapy to surgical wound closure⁶.
       A medical aspirator system was used to drain the wounds with 125mmHg continuous negative pressure and changed the dressing every 48 hours as in the other reports^2,3,7,8,10. We did not use polyurethane ether sponges in the control group because we wanted to compare NPWT with standard saline moist gauze dressing, which does not require any sponges. However, this difference in the dressing techniques is a flaw in the experimental design of this study.
       Before the dressing changes in the NPWT group, the patients were given intravenous analgesics because the polyurethane sponges were sticking to the wounds and causing pain to the patients. In the control group, the dressings were changed twice a day with less pain. Therefore, oral analgesics were used when necessary in the control group.
       In the treatment of diabetic ulcer wounds, NPWT provided a faster wound resolution compared to saline-moistened gauze. NPWT may be beneficial in diabetic nonhealing wounds of lower extremity^4–6. The surface areas of diabetic ulcers decreased by NPWT more effectively than moist gauze dressings.
       In conclusion, appropriate use of NPWT may be an alternative therapy to achieve a faster granularized wound bed in diabetic foot ulcers in preparation for other closure techniques. In this small study, the results of NPWT are hopeful. We wish to work on a larger cohort of patients for comparing the effects of NPWT on the diabetic nonhealing wounds. Further studies are needed to clarify effects and indications and to modify the technique of this alternative treatment for nonhealing wounds.