Participants. The study was conducted between May 2012 and December 2013 in the outpatient unit of the Department of Dermatology, Hospital da Santa Casa de Misericórdia de Curitiba, Pontifícia Universidade Católica do Paraná (PUC-PR), Brazil. One hundred, nine (109) patients with diabetes and foot ulcers from the Curitiba metropolitan region were screened according to the inclusion and exclusion criteria presented in Table 1. The study was approved by the Institutional Research Ethics Committee of the PUC-PR (protocol no. 22.670) and was registered in Plataforma Brazil with the National Commission of Ethics in Research (no. 0.1051212.0.0000.0020). Written informed consent was obtained from all patients before screening.
At the initial visit, a full medical history and assessment of the patients’ present condition were recorded. Study variables were collected on a paper/pencil instrument and evaluated; information included sociodemographic data, clinical history, and clinical evaluation of the wounds and amputations. The diabetic status of patients (ie, type, duration, glycemic management, including the determination of plasma glucose and glycated hemoglobin levels, current activity level, nutritional status, neuropathic pain assessment,23 and blood tests) also was recorded. Blood test results included glucose levels, glycated hemoglobin, serum albumin, blood count, erythrocyte sedimentation rate, and ulcer bacterioscopy and culture were obtained at baseline and at the study conclusion.
Treatment with C. officinalis hydroglycolic extract. All enrolled patients were treated with the C. officinalis hydroglycolic 4% extract. The extract composition included C. officinalis L. 4% and excipients (butylated hydroxytoluene, parabens, ethanol, polyethylene glycol, and purified water). The spray solution was prepared by authorized compounding pharmacies using commercial ingredients. The certificated analysis of the C. officinalis hydroglycolic extract included the following compounds from dichloromethane fraction: β-amyrin (6.7%), lupeol (4.7%), ψ-taraxasterol (8.1%), calenduladiol monoesters (5.5%), arnidiol monoesters (15.7%), faradiol monoesters (35.2%), others (24.1%), and from aqueous fraction: total flavonoids content of 120 mg/mL. Bioactivity was previously demonstrated using high-performance liquid chromatography.19,20 The DFU was cleaned twice daily with 25 mL of sterile physiological saline solution, after which 0.018 mL/cm2 of wound area of C. officinalis extract was sprayed on the wound. After allowing the solution to dry in the wound bed for 5 minutes, sterile, nonadherent, saline-moistened gauze and bandages were applied. Patients were provided cushioned footwear, diabetic boots, crutches, and wheelchairs to offload the affected areas. None of the patients used additional wound healing medication, phytopreparation, hydrogels, hydrocolloids, or supportive therapy (ie, electrotherapy, vacuum therapy, laser therapy, phototherapy). The nursing team instructed patients or their caregivers to use sterile gauze dressings after each C. officinalis extract application and to avoid bearing weight on the affected limb by using adequate footwear. Patients received their first treatment in clinic then performed care at home. The nursing team monitored whether the instructions for treatment were adequately followed through weekly phone calls to patients or their caregivers. Footwear such as cushioned shoes and diabetic boots was distributed by the local public health system and customized for the study patients.
Assessment of C. officinalis hydroglycolic extract on DFU healing. DFUs were assessed at baseline and then twice a week during visits by the nursing staff. At each visit, the DFUs were clinically assessed for appearance, size, and size reduction rate using photographs analyzed by computerized planimetry, according to a methodology described previously.24 Photographs of each ulcer were taken with a Sony DSC-H1 digital camera (Sony USA, New York, NY, USA) at every nurse visit. Image capture was standardized using a tripod frame to support the camera fixed perpendicularly to the ulcer. A circular self-adhesive label with a known area was placed close to the ulcer as a calibrator used by the software for area quantification. Digital images obtained were analyzed using Image J® software (National Institutes of Health, Bethesda, MD, USA). The software delineates the margin of each ulcer and uses different shades of color to define and calculate areas with different types of tissue.
The clinical appearance of the ulcers was assessed for different tissue types such as granulation, epithelialization, fibrin slough, and necrosis. The presence of a specific tissue type was confirmed if computerized planimetry showed a relative area >20% of the total ulcer area. The presence of exudate was confirmed by the appearance of moist gauze during dressing changes. The presence of odor was noted. The ulcer was classified according to the Wagner Grading System,25 which was used to establish DFU depth and presence of infection. Microbiological flora of the DFUs was identified using a biogram/antibiogram of a swab from the wound bed. A validated quantitative swab technique was performed to assess wound contamination and infection.26 A wound was considered infected if a high level of bacteria (1 × 106 CFU) and signs of increased erythema, exudate, odor, warmth, edema, and/or pain were present. Patients with infected wounds were treated with systemic antimicrobials; patients who had a fever and other complications from wound infections during the treatment were discontinued during the study.
Complete wound closure (ie, healing) was defined as full epithelialization of the ulcer with the absence of drainage. Patients were monitored biweekly by the study physician and trained nursing staff for 30 weeks or until healing. Healing was confirmed 1 week following closure, and the patient was monitored for another 2 weeks.
If a patient experienced an allergic reaction to the C. officinalis hydroglycolic extract, an evaluation by the principal investigator was conducted, and the patient was removed from the clinical test.
Pain assessment. Pain was assessed in 2 ways. Patients were asked to complete the paper/pencil Neuropathic Pain Scale Questionnaire (NPSQ) on the first visit (baseline) and after completion of the 30-week treatment under the guidance of the nursing staff for evaluation and discrimination of neuropathic pain, according to the methodology described previously.23 Briefly, patients were rated for numbness (no numbness sensation = 0, worst numbness imaginable = 100), tingling pain (no tingling pain = 0, worst tingling pain imaginable = 100), and increased pain due to touch (no increase at all = 0, greatest increase imaginable = 100). The NPSQ scores for these symptoms were multiplied by its specific coefficients, summed, and then subtracted by a constant to obtain the discriminant function score (DFS). DFS <0 predicted no neuropathic pain, whereas DFS ≥0 indicated neuropathic pain.
For pain related to DFU dressing changes, patients were asked to rate their pain using a numerical rating scale (NRS) that ranged from no pain (0) to the worst possible pain (10).27 The NRS questionnaire was completed at baseline and then once a month during the nursing visits.
Statistical analysis. Descriptive and analytical data were collected and stored using a Microsoft Office Access 2010 database, drawn from documents used in medical evaluations and nursing. Data were analyzed by importing Access files to the database to the Statistical Package for Social Sciences (SPSS) version 20 (SPSS, Inc, Chicago, IL, USA). All data were verified by double key entry as entered and stored in the Access database. The Access database was stored in a computer with an Intel Core 2 Duo processor, 2.2 GHz, and 2G of RAM.
Quantitative variables were analyzed using descriptive statistics. Unless otherwise indicated, data are presented as mean ± standard deviation. Qualitative variables were described as frequencies and percentages.
To evaluate the association between gender and outcomes of healing at 30 weeks, Fisher’s exact test was used. The wound contraction per week (WCw) was calculated as the baseline wound area (Ai) − final wound area (Af) ÷ the number of weeks.28
The percentage reduction in wound area per week (%RWAw) was calculated as follows:
To compare the groups defined by healing within 30 weeks, the quantitative variables were analyzed using Student’s t-test, and the nonparametric variables were analyzed using the Mann-Whitney U test. Time to complete ulcer healing was measured as the number of days from the start of treatment to the date a patient achieved complete wound closure. Wound healing was observed on a weekly basis, and the time until complete healing was estimated by calculating a cumulative frequency chart. P <0.05 was considered to be statistically significant.