Noncontact Normothermic Wound Therapy and Offloading in the Treatment of Neuropathic Foot Ulcers in Patients with Diabetes
An estimated 5% to 6% of the population of the United States has diabetes.1 A common complication of the disease is plantar ulceration secondary to sensory neuropathy and the repetitive stress of walking.2
Nonhealing foot ulcerations secondary to diabetes have been implicated in up to 90% of all amputations in Japan, Taiwan, Spain, Italy, Canada, the United States, and England.3-5 Half of all nontraumatic lower extremity amputations reported in the United States occur in people with diabetes; minorities account for a disproportionately high number of these cases.6
If clinicians are to have an impact on the diabetic population, attention must be directed to preventing ulceration or, in cases where ulceration is already present, to facilitating wound healing. In addition, glucose must be properly controlled and other factors such as adequate blood supply and pressure reduction in the lower extremities must be addressed.
Although blood supply may be adequate to support the basal needs of intact tissue in people with diabetes, the increased oxygen demands for healing are quite different.7 Without adequate blood flow and oxygenation, tissue infection rates rise and amputations become more likely. Because vascular supply to wounds is often compromised, a hypoxic state frequently exists. This, in turn, correlates with decreased tissue temperature. Typically, hypoxic wounds have temperatures between 26o to 33oC8-10; therefore, heat may be one of the most effective means of increasing oxygen perfusion in tissues and, subsequently, reducing infection.11,12
Van Hoft's law states that for every 10o-C rise in tissue temperature, a two- to threefold increase in cellular metabolism occurs.13 Excessive heat is detrimental, as thermal burns can result. The challenge is to find a safe means of providing controlled, mild warming of hypothermic tissues.
A relatively new device has been developed to deliver gentle warmth to wounds via a noncontact, semiocclusive dressing. The device, termed Warm-Up wound therapy® (Augustine Medical, Inc., Eden Prairie, Minn.), delivers humidified warmth to the wound and periwound skin at a normothermic temperature of 38oC ± 1oC. Recently published controlled clinical trials14,15 and case reports16,17 support the effectiveness of this device in accelerating healing in patients with pressure and venous insufficiency ulcers.
In addition to circulation, another factor that must be addressed when managing patients with diabetic foot ulcerations is offloading the extremity. If high plantar pressures persist, foot wounds will not heal.18 A variety of mechanisms have proven useful in decreasing plantar foot pressures. The present day gold standard is the total contact walking cast (TCC), first introduced in the United States in 1965.2 The TCC is designed to shift high pressures from wounds and bony prominences by distributing weightbearing forces over as large an area as possible. However, total contact walking casts are not always feasible when daily wound care is required. In such cases, custom fabricated posterior walking splints or halfshoes with metatarsal bars are often preferred.18
Because most noninfected, neuropathic foot ulcers heal when pressure is removed from the wound, the authors sought to determine if the healing rate could be accelerated by adding controlled warmth. A randomized, controlled clinical trial was undertaken to address this question.
Subjects and inclusion criteria. Following Institutional Review Board approval, patients were selected to participate in the study based upon a history of a nonhealing diabetic foot ulceration. Patients with active cellulitis, purulence, fever, osteomyelitis, or those determined by the referring physician to have inadequate blood supply to support healing were excluded. Thirty-six subjects completed the study (18 treatment/18 control).
Initial assignment to the control or treatment groups was determined by a toss of a coin. The initial selection resulted in the first subject being assigned to the treatment group. Following this, each subsequent individual referred to the study was assigned in an alternating fashion.
Inclusion in the study required that the participant had a wound on the leg over a bony prominence and appeared secondary to pressure. Of the individuals assigned to the experimental group, 67% (12) had wounds on the plantar aspect of the toes or metatarsal heads, compared to 78% (14) of the individuals in the control group. Other wound locations included two malleolar, one plantar midfoot, and seven calcaneal (equally divided between the groups).
Protocol. Following explanation of the study and signing of informed consent, a tracing of wound surface area was made onto clear acetate film with a fine-tipped indelible marker. Wound surface area and perimeter were calculated by computer digitization of the acetate tracing (Bioquant, R&M Biometrics, Inc., Nashville, Tenn.). Patients in the control group had their wounds cleansed and dressed with an appropriate moisture-retentive dressing and received an offloading device that was applied by the nursing staff. The dressings most frequently used were calcium alginates combined with thin, semipermeable foams or semipermeable foams alone, depending on wound hydration. Offloading was accomplished using an OrthoWedge (Darco International, Inc., Huntington, WV) cast shoe with a metatarsal bar. In cases where foot deformity or mid- and hindfoot wounds were present that could not be effectively offloaded by such a device, a customized posterior walking splint was applied. The patient was instructed in daily wound care, including saline irrigation and dressing application. Even though daily dressing changes may not have been required, subjects in the control group still were instructed to change their dressings daily because those in the treatment group would be undergoing daily dressing changes in order to receive the warming intervention. Subjects also were instructed to maintain daily blood glucose logs as directed by their physicians in the diabetic foot clinic.
Individuals in the treatment group had their wounds measured and documented in a similar fashion. They also were instructed in the application and use of the warming device. The warming system is comprised of a noncontact foam dressing, warming card, temperature control unit, and AC adapter (see Figure 1). Subjects were told to cleanse their wounds with saline, then apply the noncontact wound cover. Depending on wound location, one of two types of wound covers was used. The standard type dressing was a 2-cm x 2-cm absorbent foam ring, 1-cm thick, with an adhesive border. The center portion of the ring was covered with two layers of clear plastic that secured the warming card and also permitted visualization of the wound. In the case of toe wounds, a "bag-type" wound cover (see Figure 2) was used. This wound cover basically enclosed the distal foot in a plastic sleeve into which a 2-cm x 2-cm dressing had been placed. When the wound cover was in place, subjects inserted the metallic warming element between the two layers of the plastic window (see Figure 3). The warming element was attached to the temperature control unit and the unit was turned on. Treatments were conducted daily for 3 hours, 5 days per week. During the time of application, the patient turned on temperature control unit for 1 hour, then turned off the unit for 1 hour, but the noncontact dressing was left in place. Finally, the unit was turned back on for an additional hour to complete the 3-hour treatment. The researchers set the time protocol because one of the goals of the study was to determine whether limited use of the warming device would be beneficial. If beneficial effects could occur with limited use, this would (hopefully) make the unit more acceptable to patients.
At the end of the 3-hour period, patients in the treatment group dressed their wounds with alginate and semipermeable foam dressings, or semipermeable foam dressings alone, as directed by the investigator. They applied offloading devices in the same manner as the control subjects and likewise maintained a daily glucose monitoring log.
All experimental and control subjects returned to the clinic on a weekly basis for wound assessment. Transportation costs were reimbursed, which increased compliance. In the rare instances where subjects could not keep their scheduled appointments, they returned within 2 days of the appointed time. At these weekly visits, wounds were irrigated with saline pulsatile lavage with suction by means of the Simpulse Varicare System® (Davol, Inc., Cranston, RI) and any remaining necrotic tissue was debrided with a scalpel. Surface area tracings were made, general wound characteristics were documented, and average blood glucose measurements were noted. Subjects continued in the study for 2 months or until the wound healed.
Analysis. The Gilman equation was used to standardize healing rate measurements.19 The equation takes into account not only wound area but also the wound perimeter; thus, allowing an unbiased measure of the linear advance of the wound margins toward the wound center. The equation, as stated by Gilman, is: healing rate (eg, cm2/day) = [2 x (Aa - Ab)] / [(Pa + Pb) x (b-a)]. This formula looks at the difference in surface areas from the start to the end of treatment, the average perimeter of the wound at the same two points, and the time interval between measurements.
After testing for normality, a Student's t-test was used to evaluate for differences in mean values of the two groups based on wound starting size, subject age, and average blood glucose. A Student's t-test also was used to evaluate differences in length of time in study and time to healing for the two groups.
A comparison of the standardized healing rates of the subjects in the two groups was performed by use of a one-way analysis of variance (ANOVA). Additionally, a chi-square analysis was performed to evaluate the total number of participants that healed in each group. P values of < 0.05 were considered statistically significant for all tests.
The average age of patients in the control group was 52.5 (± 12.1) and no significant differences were shown between patient groups with respect to baseline wound size and blood glucose levels (see Table 1).
The treatment group had a mean healing rate of 0.019 + 0.019 cm2day compared to a mean healing rate in the control group of 0.008 + 0.009 cm2/day. The difference in the means of the two groups was statistically significant (F = 4.18; df = 1,34; P = 0.049) (see Figure 4).
Of the 18 subjects in the warmth-treated group, 13 (72%) healed within the 2-month timeframe of this study. Five (28%) of the 18 subjects in the control group who achieved complete healing. The difference is statistically significant between the groups (chi-squared = 9.03; df = 1; P = 0.003).
The length of time in the study was comparable for both groups, as was the average time to healing for subjects that achieved complete wound closure within the study time frame. The average length of time in the study for treatment subjects was 41 ± 20 days and 52 ± 17 days for the control group. This difference was not statistically significant (t = -1.79; df = 34; P = 0.082). Eighteen patients in the treatment group and five in the control group achieved complete closure; the average time to closure was 32.6 ± 17.1 days and 27.6 ± 13.7 days, respectively. These differences were not statistically significant (t = 0.58; df = 34; P= 0.57).
The patient with complications secondary to diabetes poses unique challenges for the wound care practitioner. Peripheral neuropathy leaves the patient with autonomic, sensory, and motor changes. One might surmise that loss of vasoconstrictive tone would lead to increased blood flow and a concomitant increase in tissue Po2. Although true to some degree, finding tissue ischemia around diabetic wounds is not uncommon, because large and small vessel disease do not progress at the same rate.20 Evidence of small vessel disease is present in the form of trophic changes in the skin and adnexal structures. Skin dryness and fissuring are common and, in the presence of sensory and motor neuropathy, lead to high pressure areas over bony prominences and ultimate tissue failure. Local warmth and a humid environment, as provided by the warming device, appear desirable if healing is to be promoted, but provide only a partial solution because pressure also must be alleviated.
Results of a meta-analysis of studies involving 586 patients with diabetic foot ulcers receiving "good wound care" showed that 16% of ulcers healed in 8 weeks. "Good wound care" was defined as saline moistened gauze or placebo gel and gauze with the addition of instructions in the importance of avoiding weight bearing.21,22 Some patients were provided with offloading devices and others just were told to remain nonweightbearing. Total contact casting was not utilized. The difference between these protocols of care and those used by the authors (local wound care with calcium alginate and/or foam dressings and offloading) may explain why patients in the authors' control group also had better outcomes than those reported in the literature.
Total contact casting can effectively reduce pressure and accelerate healing. Armstrong et al 23 reported a mean duration of casting to healing in 25 people with diabetes to be 38.8 ± 21.3 days. When pressure exceeded 99 N/cm2, the mean days to healing rose to 53.4 ± 31.4. In the study presented here, patients who received warming therapy in addition to offloading healed at a faster rate than those who received only offloading. Among the 13 patients in the treatment group who healed within the 8-week period, the days to healing averaged 32.6 ± 17.1, which is a better time than reported in the Armstrong study.23 These findings seem to reinforce the accepted teaching that offloading must be an integral component of diabetic foot wound treatment.
A limitation of the warming device is that it does not permit ambulation. Even though patients reported that their treatments occurred in the evening while watching television and appropriate offloading of both groups should have minimized stress on the wounds, this may have affected the results.
Elevated blood glucose values also retard wound healing. Average blood sugars (mg/dL) between 144 and 306 are associated with a high risk of complication such as poor wound healing.24 Subjects in both the treatment and control group were from a high-risk population and many had poor glucose control. Lower glucose values could have facilitated more rapid wound healing. Even though the treatment and control group did not differ significantly in this respect, glycosylated hemoglobin measures would have provided better information on glucose values over the course of the study.
Because warming therapy appears to offer a valuable means of accelerating wound healing, efforts should be directed toward finding ways to incorporate such therapy into an ambulatory offloading apparatus. Without this adaptation, the tendency for individuals to ambulate without protective footwear is increased. In addition, the fact that an individual must remove the noncontact dressing and apply another wound care dressing in order to ambulate poses a two-fold problem. First, much of the benefit of the noncontact dressing in maintaining a warm, moist environment where migrating epidermal cells are not disturbed, is lost. Second, having to apply another wound dressing after warming treatment significantly increases the cost of care. The noncontact dressing is designed for long-term use and appears the most desirable. Furthermore, having the ability to provide continual low-level warmth throughout the day may likely prove more beneficial so the wound will not experience sudden changes in temperature.
Warming therapy appears to be a logical approach to accelerating wound healing. The challenge is to develop a delivery process for the modality that combines a noncontact moist wound environment with a device that permits ambulation.
This study was funded by a grant from Augustine Medical, Inc. Eden Prairie, Minn.