Efficacy of Polyurethane Foam Dressing in Debrided Diabetic Lower Limb Wounds
Currently in India, there are 30 million people with diabetes and an estimated 40,000 amputations occurring annually due to diabetes-related foot problems. The most common cause is the infected neuropathic foot, which is potentially preventable.1 Few organized diabetic foot screening programs exist in India. Podiatric services are available only in major centers in India, and a multidisciplinary team approach is mostly lacking. Proper orthotics for patients with diabetes are not readily available. Socio-economic factors, such as barefoot walking, inappropriate footwear usage, lack of awareness of the seriousness of diabetic foot problems among doctors and patients, and hence, late referrals to specialty centers, are a matter of concern. Seventy percent of India’s population lives in the rural area and 40% stay in 1-room tenements. Inadequate sanitation, improper foot offloading due to lack of facilities, poor socio-economic conditions, and inadequate awareness about the seriousness of diabetic foot problems are common. Rat and insect bites, vigorous massage, thermal injuries due to hot fomentation, injuries due to improper footwear, and fungal infections of intertriginous skin cause a significant number of diabetes-related foot injuries in India.2 These cause extensive necrotizing fasciitis and other soft tissue and bone infections of the lower limb, which can be limb and life threatening.3 In India, few patients are insured, and the vast majority of patients with diabetes and foot problems have to pay for the cost of medical care. Hence, the cost of treatment and consumables used assumes much greater significance in such circumstances.
In the proliferative phase of wound healing, which lasts from 5 days to 3 weeks after injury or trauma, fibroblasts migrate in the wound depths. These fibroblasts synthesize and secrete collagen. Their migration is self-powered and limited by contact inhibition. Fibroblasts do not contain fibrinolytic enzymes, and the fibrin, dead cells, and tissues in a wound can inhibit their migration. Polyurethane foam dressings facilitate faster removal of slough and dead tissue and assist in this stage of wound healing and in epidermal migration. Polyurethane foam dressings loosen slough by creating a moist wound environment, assist in proper wound bed preparation, and promote this phase of wound healing.4 Slough is a complex mixture of deoxyribonucleic proteins, fibrin, bacteria, leukocytes, and serous exudate. Slough is not dead tissue. Wound exudate can be classified by quantity as mild (0.25 g/cm2/24 hours), moderate (0.5 g/cm2/24 hours), and heavy (1.0 g/cm2/24 hours).5 Medicated polyurethane foam impregnated with silver, iodine, alginate, and various other substances is widely used in the management of diabetic wounds.5 Few studies regarding the efficacy of non-medicated polyurethane foam dressings have been conducted. Polyurethane foam dressings retain moisture, maintaining a moist wound environment, which is important for proper wound healing. They also absorb excessive exudate thereby preventing tissue maceration, facilitating removal of slough, and promoting the proliferative stages of wound healing.6 Medicated polyurethane foam dressings, though effective, are costly. Non-medicated polyurethane foam is inexpensive, readily available, and sterilizable, making it a cost-effective dressing material. It gives sufficient mechanical protection to the wound, is nonadherent and nonallergenic, and does not shed loose material into the wound. Moreover, it conforms to anatomical contours and has a long shelf life.7
Materials and Methods
This prospective study was conducted between January 1, 2005, and July 31, 2005, at the Amrita Institute of Medical Sciences and Research Center at Kochi, in the south Indian state of Kerala. This facility is a super-specialty, 800-bed tertiary referral hospital. The Division of Diabetic Foot Surgery is associated with the Department of Endocrinology. Vascular surgery, radiology, reconstructive surgery, and orthopedics are all under one roof. The diabetic foot division has a daily average outpatient load of 40 and an average of 45 inpatients daily.
Patients with type 2 diabetes who presented with lower limb wounds underwent adequate surgical debridement under anesthesia (ankle, popliteal, femoral, or sciatic blocks; and when needed, spinal or general anesthesia) for necrotizing fasciitis and other soft tissue infections of the lower limb. All necrotic tissue and slough were excised in the operating room. Patients with wounds that were detected as having reformation of slough with or without excessive exudate in the immediate postoperative period (within 72 hours) were included in the study. Patients were then randomly assigned to either the study group or control group. The remaining patients in whom the wounds did not reform slough and remained clean in the immediate postoperative period were excluded from the study.
In the study group, non-medicated polyurethane foam (industrial-grade foam) was used as the dressing material. Each polyurethane foam sheet measured 10 mm x 10 cm x 30 cm. It had a pore size of 0.4 mm diameter (65 pores per square inch). Poisson’s ratio (ie, the ratio of transverse strain to longitudinal strain during longitudinal stretching) was measured at 0.03 g/cm3. It had a Shore hardness of 10. Shore hardness denotes the density of the material and its resistance to indentation from a spring-loaded indenter. A higher Shore number indicates that a material has greater density. Dry weight of each sheet was 7.7 g; weight when fully soaked with water was 156 g (20.25 times the dry weight); and the weight of the dressed foam sheet was 73.3 g (9.5 times the dry weight).
The wounds were initially cleaned with sterile normal saline solution; bedside sharp debridement of necrotic tissue and slough was performed when required; and the wound was again cleaned with normal saline. The polyurethane foam sheet was soaked in sterile saline and then manually squeezed to remove as much saline as possible. This procedure was completed to make the foam more conformable. The foam sheet was then placed directly on the wound surface. Sterile gamgee pads (cotton pads with gauze on both sides) were placed over the foam sheet and held in place with gauze bandage that was secured with a light compression elastocrepe bandage applied with 50% overlap. In some cases in the study group, polyurethane foam dressings were continued post-operatively when a patient required re-debridement in the operating room. No topical antibiotics, desloughing ointments, or other topical agents were used in the study group.
In the control group, patients underwent surgical debridement in the operating room where infected tissue was excised under anesthesia. Regular bedside sharp debridement was performed, and wounds were dressed daily with conventional techniques using topical antibiotics, desloughing agents (eg, collagenase, papain-urea, hyaluronidase ointments), or hydrogel and hydrocolloid dressings as deemed necessary, taking into account the wound status. In the immediate postoperative period, the affected limb was strictly offloaded in both groups.
Patients were subsequently mobilized on crutches or walkers to offload the affected limb until wound healing. Patients of both groups were administered culture-specific parenteral antibiotics based on the deep tissue culture sensitivity reports taken at the time of surgery. Insulin was used to tightly control diabetes in both groups.
Split skin grafting (SSG) was performed in the study and control groups for all wounds > 5 cm (approximate surface area of 20 cm2) once the wounds were found to be clean, devoid of slough, and granulating well and culture of the surface swab did not show any significant bacterial growth. In both groups, wounds < 5 cm in diameter (approximate surface area of 20 cm2) were allowed to epithelize and heal without SSG.
In both groups, patients that did not undergo SSG upon initial admission were continued on dressings after discharge, where the bystander (eg, family member, home nurse, caretaker) was trained in the specific dressing techniques. Weekly or bi-weekly review in the outpatient department was advised, according to the condition of the patient. The patient was readmitted when the wound was ready for SSG. All patients were followed up for a minimum of 3 months. In both groups, peripheral occlusive vascular disease was assessed by an ankle brachial index (ABI). Vascular insufficiency was diagnosed when ABI was 0.8 or less. A vibration perception tester (Biothesiometer, Bharat Biotech, Bangalore, India) was used to assess peripheral neuropathy. Neuropathy was diagnosed in all patients with values of 15 V or more. Routine laboratory investigations for blood sugar levels, blood urea nitrogen, serum creatinine, and white blood cell count were also documented. The site of ulceration and size of the wounds immediately after debridement were also noted.
The endpoint was the number of days from the date of initial debridement to the date of the SSG in patients who underwent SSG. In patients whose wounds did not require SSG, the endpoint was the number of days from the date of first debridement to the date when the wound was considered completely epithelized. The wound was categorized as “not healed” if there was incomplete epithelization of the wound after a minimum period of 3 months follow-up after debridement.
In both groups, peripheral angioplasty was used in the revascularization of patients with peripheral occlusive vascular disease.
Statistical analysis was performed using SPSS software (SPSS, Chicago, Ill) and the statistical methods followed for detecting the P value were executed using the chi-square method. The Kaplan-Meier graph depicts the probability of wound healing for a specific number of days after surgical debridement.
The study included 48 patients with type 2 diabetes. The study group and the control group each had 24 patients. There were a total of 28 men and 20 women in the study. The mean age of the patients was 58.8 ± 9.4 years in the study group and 52.4 ± 7.4 years in the control group. The mean duration of diabetes was 14 ± 8 years in the study group and 13 ± 7 years in the control group (Table 1). At the end of the 3-month follow-up period, wounds of 7 control group patients (29.2%) had not healed while all wounds in the study group had healed (P = 0.014). The mean number of days taken for wound healing in the study group was 22.5 ± 15.4 days (24/24 patients), while in the control group it was 52.0 ± 22.7 days (17/24 patients, P < 0.0001). The Kaplan-Meier graph shows the median time to wound healing was 16 days in the study group and 60 days in the control group (Figure 1).
The number of patients who underwent SSG was 19/24 (79.2%) in the study group and 12/24 (50%) in the control group (P = 0.034). All wounds > 5 cm in diameter (approximate surface area of 20 cm2) underwent SSG. After surgery, exposed bone was evident in 16 study group patients (66.7%) and 12 control group patients (50%, P = 0.475). Peripheral occlusive vascular disease was diagnosed in 8 study group patients (34.8%) and 9 control group patients (37.5%). Mean blood urea was 44.6 ± 28.3 mg% in the study group and 49.7± 43.2 mg% in the control group. Mean serum creatinine was 1.7 ± 1.4 mg% in the study group and 1.4 ± 0.8 mg% in the control group. White blood cell count was 23.2 ± 8.5 cells/mm3 in the study group and 21.4 ± 7.2 cells/mm3 in the control group (Table 1). As per site of infection, the left leg was involved in 2 study group patients and 3 control group patients; the left foot in 12 study group patients and 10 control group patients; the right leg in 6 study group patients and 1 control group patient; the right foot in 2 study group patients and 8 control group patients. Four patients had thigh wounds. Mean surface area of the wounds was 208.9 ± 196.3 cm2 in the study group and 198.3 ± 186.8 cm2 in the control group. Neuropathy was present in 6/24 control group patients and 15/24 study group patients (P = 0.526).
The state of the wound bed, the phase of wound healing, the amount of wound exudate, bacterial load, and host resistance all play an important role in determining the most appropriate dressing material to be used in patients with infected wounds. The type and technique of the dressing will vary accordingly. In a review article, Kirsner et al8 demonstrated that sustained-release silver foam dressings hastened granulation formation compared to plain foam dressings over a period of 4 weeks. The authors described the use of sustained release silver foam dressings in the treatment of nonhealing chronic wounds and showed that they significantly decreased odor and exudate faster than nonactive foam dressings. The results showed a 45% reduction in ulcer size from the baseline for the sustained release foam dressing compared with 25% for the foam dressing without silver. Results also showed a 50% reduction in ulcer size for the sustained release foam dressing compared with 30% for local best practice.
Other studies have not shown any significant difference in healing time when using gauze, foam, and alginates.9,10 Comparison studies of the efficacy of different hydrocolloids and foam dressings have found no significant difference in their effectiveness on wound healing.11
In the present study, wounds dressed in sterile, non-medicated, polyurethane foam granulated earlier and epithelized faster compared to wounds dressed with conventional dressing methods (P < 0.0001). This occurred because the foam creates an ideal microclimate for wound healing as it maintains a moist wound environment.9 As a result, the slough and necrotic tissue loosen, and wound bed preparation becomes easier. More complete slough excision can be achieved at each bedside debridement.12 The sharp debridement was also relatively pain free. The foam was found to be relatively nonadherent and did not directly cause substantial mechanical debridement at the time of dressing change. Due to early granulation formation, the study group was grafted earlier as compared to the control group. Even in patients with a wound surface area < 20 cm2 and that did not require SSG, epithelization proceeded faster. Though there was no significant difference in the mean surface area of the wounds between the 2 groups, more patients in the study group had wounds with a surface area > 20 cm2 thus requiring SSG (P = 0.034).
Another important quality of the foam is its capacity to absorb excessive exudate from the wound. Wound exudate occurs due to bacterial infection, and the amount of wound exudate will be directly proportional to the degree of bacterial colonization.12 Excessive exudate, if allowed to remain on the wound bed, will cause tissue maceration, which will promote slough formation and delay wound healing.6,12 Absorbent dressings, including polyurethane foam, will prevent this problematic situation. The fluid absorption capacity of the foam will depend on its Shore hardness, pore size, and number of pores per square inch. A higher Shore number indicates that a material has greater density.13 By testing foam of different Shore hardness, the authors determined that the foam with a Shore hardness of 10, a pore size diameter of 0.4 mm, and 65 pores per square inch was ideally suited for the management of wounds in this study. Although wound exudate was not estimated quantitatively, most wounds were clinically classified as having moderate exudate.
To test the absorbency of the polyurethane foam, the dry weight of the sheet was noted, then the sheet was completely soaked in water. After the dripping had stopped, the foam was reweighed to calculate total absorbency (20.25 times the dry weight). The foam sheet was then wrapped on an acrylic limb model, covered adequately with gamgee pads, and dressed with gauze and elastocrepe bandages. After 30 minutes, the sheet was weighed again to measure the absorbency of the polyurethane foam when dressed (9.5 times the dry weight).
The authors noted that in the study group, marginally more patients underwent deeper wound debridement in the operating room, where infected fascia, tissue, and muscle were excised, leading to postoperative bone exposure (P = 0.475). No significant difference was seen between the 2 groups regarding peripheral obstructive vascular disease and laboratory investigation results (ie, white blood cell count, blood urea nitrogen, serum creatinine, and blood sugar values). However, neuropathy was evident in more control group patients compared to the study group patients (P = 0.526).
The polyurethane foam utilized in the study is generally used for upholstery and packing. This is readily available and inexpensive. One roll of cleaned and autoclaved non-medicated polyurethane foam pack (30 cm x 10 cm x 10 mm) costs the patient $0.42 (US) or Rs 20/- (Indian Rupee), while sterile medicated packs of 10 cm x 10 cm foam cost the patient $5.77 and Rs 275. Another branded product noted was sterile Lyofoam (ConvaTec, Princeton, NJ), which measured 17 cm x 10 cm and cost $10.72, Rs 495/-; at 20 cm x 15 cm, it cost $13.60, Rs 625/-.14 Not only did the study foam cost less than available marketed products, it also promoted faster healing. This translated into reduced costs in terms of shortened hospitalization and more rapid return to work, thus reducing the overall direct and indirect costs of treating the diabetic foot wounds. Any reduction in cost is particularly significant in India, since the vast majority of patients have difficulty meeting the cost of treatment because they do not have medical insurance.
The results suggest that there was significant reduction in the time taken for diabetic lower limb wounds with slough and exudate to heal when sterile, non-medicated polyurethane foam dressings were used as compared to conventional dressings. Also, the direct cost of the polyurethane foam used in the study is far less than marketed dressings, and it is readily available locally.