Introduction

Elective and emergency arterial surgery for lower limb revascularization (rearterialization) is a common, effective procedure for treating Critical Limb Ischemia (CLI). Despite surgery and successful revascularization, chronic ischaemic leg ulcers in CLI can take weeks or months to heal, and therefore, these ulcers are a challenge also for postoperative ulcer care [1,2]. In patients with CLI and poor peripheral circulation, the operation wounds are often complicated, which results in bacterial infections at the Surgical Site (SSI) [3-7]. Conditions that give rise to complicated SSI wounds and chronic ulcers in CLI are, however, multifactorial. A number of co-morbidities can contribute to complications and chronicity of the ulcers, including diabetes, chronic venous insufficiency, malnutrition, smoking, hyperglycaemia, older age, anemia, renal, cardiac or hepatic failure, malignancies, and the use of some drugs, like corticosteroids or immunosuppressives [8- 10].The incidence of SSIs varies after peripheral vascular surgery from 4% to 27% in patients with CLI [11], Like a slow healing of chronic ulcer in CLI, the healing of ulcers with SSI might be delayed in approximately 20% of patients after peripheral arterial revascularization, despite the restoration of perfusion and oxygen supply [1,2,10,12,13].

After revascularization surgery, conditions are not favourable for additional major surgical procedures; therefore, patients are typically referred postoperatively to topical wound care [14,15]. Although hundreds of wound care products and protocols are available, evidence remains tenuous of the effectiveness of different postoperative treatment strategies. Prospective trials on treatments success for CLI ulcers are also rare [16]. In recent years and due to the risk of appearance of resistant microbe strains, it has become important to find new antiseptic tools and safe natural alternatives for long-term topical wound care, instead of use of antibiotics. One is a resin prepared from natural coniferous Norway spruce (Picea abies) [17]. The other is refined medical grade honey, derived from the Manuka myrtle (Leptospermum scoparium) [18]. Both have been used for treating chronic skin wounds with or without clinical signs of infection [17-23]. The objective of this 6-month prospective, clinical study was to investigate the healing rate of chronic leg ulcers in patients that underwent peripheral arterial revascularization for CLI and who received post-surgical topical ulcer (wound) care with two optional antiseptic salves. The study aimed to determine the frequency of ulcer healing after revascularization surgery and discover whether the antiseptic resin and honey salves are effective and safe alternatives as topical postoperative ulcer care agents at home. We also aimed to provide preliminary data on whether resin and honey salves show any marked differences in efficacy as topical agents for ulcer healing in patients with CLI that had undergone the revascularization.

Materials and Methods

Study population

The study population consisted of 40 consecutive patients with chronic ischaemic ulcers and CLI with or without preoperative chronic leg ulcer. All patients included had stage IV CLI, according to the Fontane classification of the European Society of Cardiovascular Surgery, or stage 5-6 CLI, according to the American Rutherford classification [22]. All patients underwent a clinically successful acute or elective rearterialization (see operation types and patient characteristics in Tables 1 and 2). All patients were considered eligible for postoperative topical ulcer care at home if they had a preoperative leg ulcer. By Fontane stage IV or Rutherford stage 5-6 CLI the leg ulcers are ischemic, necrotic, or gangrenous, and accompanied by diffuse pedal ischemia [22]. After study inclusion, the patients with preoperative leg ulcers (35 cases) were randomized to receive either a resin or honey salve for postoperative topical wound care at home. There also were patients (5 cases) who developed after surgery a complicated and infected wound at the Surgical Site (SSI). Also, these patients with SSI were similarly treated, randomized to the salve groups and followed-up postoperatively as the patients with a preoperative leg ulcer. Demographic data on patients in the resin and honey treatment groups are presented in Table 1. The operative procedures are given in Table 2. The majority of patients were older males, often with type 1 or 2 diabetes and a smoking history, as is typically the case in patients with CLI.

Estimation of study population size

The sizes (power analysis) of the resin and honey salve groups were estimated by a statistician. The endpoint was wound or ulcer healing within 6 months after surgery. We assumed that no patients would drop out of the study. The analysis was based on a binary outcome, head-to-head, superiority trial design. We used a two group chi-squared test (X² test) with a 0.050 two-sided significance level and 80% power for observing the difference between a Group 1 proportion, p₁, of 0.900 and a Group 2 proportion, p₂, of 0.500 (odds ratio 0.111). The required sample size was estimated to be 20 patients in each group. The p₁ and p₂ values were arbitrarily chosen to exclude or include a marked, major difference in efficacy (90% versus 50%) between the test items in wound care. Simultaneously, we aimed to obtain moderately small-sized test groups. We assumed that this preliminary investigation would provide a more solid basis for calculating group sizes in future, larger investigations.

Inclusion criteria

The study population included patients that had undergone a surgical or percutaneous arterial revascularization for CLI, under either peripheral elective or emergency conditions, at the Clinic of Surgery in Kuopio University Hospital, Kuopio, Finland, between April 2014 and January 2015. Patients had a lower-limb (leg/foot) chronic ulcer of ischemic origin that was (a) infected or non-infected, but present prior to revascularization (35 patients, pre-surgery ulcer), or (b) postoperative wounds that were infected but appeared after surgery as complicated surgical chronic wounds that failed to heal postoperatively (i.e., SSI; 5 patients, post-surgery ulcer). In all cases, senior vascular surgeons considered that, after the operation and hospital stay, all ulcers required effective, active, postoperative topical care and follow-up for surveillance and control in the outpatient department. Exclusion criteria were a life expectancy of less than six months, advanced malignant disease, or an expected requirement of additional surgical revisions or plastic surgery, including minor skin transplantations. Patients were also excluded when they had a known hypersensitivity to resin or honey. When necessary, appropriate epicutaneous tests were arranged to detect the actual cause of suspected allergic reactions. All pre-surgery and post-surgery ulcers were considered to represent typical CLI-related leg wounds or ulcers that arose due to severe circulation failure in the lower limb. They were considered to require long-term, postoperative topical wound care, including intense, active postoperative attention and monitoring. After recruitment and randomization, we found that, of the 5 patients with SSIs, 4 were in the resin group and 1 was in the honey group. The average ulcer sizes were measured at the time of recruitment, after the operation procedures had been carried out (Table 2).

Randomization

Randomization to resin or honey salve group was performed by a statistician. Patient names were placed in permuted blocks of four with an allocation ratio of 1:1. The responsible physician allocated the patients to treatment groups, according to the randomization list (enclosed in envelopes). The treatments could not be blinded, because the resin and honey had discernible properties that could not be masked (e.g., color, fragrance, and consistency).

Study design, purpose of the study, and follow-up

The study group allocations and follow-up times are shown in Figure 1. The primary objective was to analyse the rate (frequency) of leg ulcer healing in the resin and honey groups, within the follow-up period of 6 months. All patients were scheduled to receive 5-7 follow-up examinations in the hospital. No significant differences between study groups were found in the distribution of completed follow-up visits or in demographic characteristics (Tables 1 and 2). Demographics, disease history, and follow-up data were recorded on clinical report forms by the responsible physician (TA). Forms were completed at the time of recruitment and at every follow-up visit throughout the 6-month period. All patients were advised to visit the surgical outpatient department, when clinically indicated and whenever the patients considered a visit justifiable. When the follow-up examination concluded that ulcer had clinically healed, the date of the last clinical report form was taken to indicate the time required for clinical healing of the ulcer (i.e., the primary objective was achieved). When the ulcer was not fully healed within 6 months at the last available followup visit, the treatment was considered clinically unsuccessful (i.e., the primary objective was not achieved). In those cases, a post hoc analysis of ulcer size was carried out with statistical linear modelling, when possible, and when sufficient follow-up data of ulcer size and follow-up time were available. The analyses aimed to obtain objective and statistically significant evidence of whether the ulcer was at a proper trend to heal, and to statistically (mathematically) estimate and predict the time point for final healing. Photographs of ulcers were taken at follow-up visits, when possible, but at least at the study initiation and at the study end (Figure 2). Any notable improvement, deterioration, or any other factor that might contribute to healing during the followup period (e.g., mechanical wound debridement, cleansing, or antibiotic treatment) was recorded on the clinical report form.

The topical treatments were started in the hospital and continued at home (self-care), according to instructions from a specialized nurse and the study protocol. Patients were instructed and prescheduled to contact the clinic at intervals of approximately 4 weeks, but at least whenever observable changes appeared, either in ulcer status or in the general health of the patient. The resin or honey salve was the only topical ulcer care remedy allowed at home. Systemic antibiotics were given only during the hospital stay, when necessary, based on clinical grounds and at the discretion of the surgeon, but not during the home care period.

Data collection and statistics

Data collection, analyses, and results reporting were conducted according to the CONSORT-statement. Differences between parallel treatment groups were compared with the c²- test or Fisher’s exact test, as appropriate. Means and standard deviations were computed for continuous variables. In calculations of percentages, the means with 95% confidence intervals (95% CIs) were estimated. After determining the data distribution, proportions were compared with either the non-parametric MannWhitney U-test and log-rank test, with the parametric Student’s t-test, or with the Z-test, as appropriate. The rates of ulcer healing were based on cumulative data. Assessments of mean and median healing times were performed with cohort tables of patients with healed ulcers in the two treatment groups over the treatment course. P-values <0.05 were considered statistically significant. Statistical calculations were performed with SPSS 20.0 (IBM Corp, Armonk, NY) and XLSTAT 2018.6 Excel software, iMac version (Addinsoft, Boston, USA).

Life table analyses

Life table analyses of the probability of ulcer healing were performed with XLSTAT 2018.6 Excel software (Addinsoft, Boston, USA), iMac version. The probability distribution function for ulcer healing over time was computed for patients in both the resin and honey groups. The observed time of clinical ulcer healing (cases in Figure 3), or the estimated healing time based on the linear model (cases in Figure 4), were used as end points of a favourable, successful event, in life-table computations. The log-rank test value was calculated to study the difference between study groups.

Statistical analyses of ulcer size

In 9 patients, the ulcers showed clinical signs of complete healing during the 6-month follow-up (Figure 3). In 13 patients, the ulcers had not healed, but follow-up data on ulcer size were available for at least three postoperative time points (Figure 4). A high number (17 cases) of patients had inadequate follow-up data (Figure 5); i.e., the ulcer had not healed and follow-up data were not complete or satisfactory for post hoc regression analysis. Furthermore, one male patient died after surgery but before entering the study. Therefore, we decided to extend the analysis of ulcer healing with a post hoc statistical approach, particularly for the 13 patients with ulcers that had not shown clinical evidence of complete healing at the last follow-up visit. In a linear analysis, statistically significant changes in ulcer size over time were considered objective predictors of the biological fate and course of the ulcer. That is, size changes indicated whether an ulcer was properly and progressively healing, even in cases where the ulcer had not completely closed (epithelized) at the last observation point.

Linear regression analysis

Ulcer size was calculated as the greatest length ´ the greatest width of the ulcer (i.e., surface area, mm²). In linear regression analyses, the ulcer size and the follow-up time (days) were used as test parameters. The regression analysis was computed with commercial XLSTAT 2018.6 Excel software (Addinsoft, Boston, USA) in an iMac computer. We analysed data from all patients with healed ulcers (Figure 3) and from patients with follow-up data on unhealed ulcers, measured in at least three consecutive follow-up visits (Figure 4). In linear regression analyses, the variance tables disclosed R², the coefficient of determination, which indicated how well the variation of the dependent variable (ulcer size) could be explained by the explanatory variable (time). The closer the R² was to 1, the better the follow-up time could explain the changes in ulcer size; i.e., ulcer was at objective trend to heal. In linear regression, three parameters are computed: the F statistic, the value of Pr>F (i.e., indicating the significance [P] of the fit between the model and the observed data), and the regression coefficient. A negative regression coefficient indicates that an ulcer has progressively decreased in size, i.e., at trend to heal. In addition, the regression model allowed to estimate and predict the time to complete healing; i.e., the time to an ulcer size of 0 mm².

Study endpoints and criteria for successful ulcer healing in mathematical modelling

The primary study endpoint was total (complete) healing and “epithelization” of the ulcer or wound. This diagnosis was based on clinical examinations and evaluations of ulcer appearance by expert physicians at follow-up visits in the hospital. Examples of healed and unhealed chronic ulcers are shown in Figure 2. The regression analysis results for healed and unhealed cases are shown in Figures 3 and 4, respectively. The post hoc regression analysis was particularly focused on the 13 cases (Figure 4) of ulcers that were not fully healed clinically, but with reliable data on ulcer size from at least three follow-up visits. In these patients, ulcers were considered to be on a successful and appropriate healing course, when the linear regression analyses between ulcer size and followup time gave a negative regression coefficient, the correlation between ulcer size and time was high (³0.80), and the coefficient of determination (R²) was ³0.80 and significant (Pr>F value £0.30). In these cases, the modelling also enabled the prediction of the healing time even though a proper healing of ulcer was not yet achieved at the last available follow-up time point.

Salves for wound care at home

Resin salve: In previous studies, the natural spruce resin salve is shown to exhibit substantial antimicrobial (antiseptic) properties in vitro, and additionally, it seems to enhance skin regeneration and wound healing in vivo, according to observations from clinical trials in patients with pressure ulcers or complicated surgical wounds [17,23]. Resin salve is a 10% (w/w) mixture of coniferous purified spruce (Picea abies) resin suspended in a standard salve base ingredient. The coniferous resin (active ingredient) was collected manually from Finnish Lapland, mechanically purified, chemically unmodified, and mixed into salve form, in accordance with GMP standards. The product carried the CE mark for approved medical devices, and it is sold in pharmacies in Europe (Abilar^® 10% Resin Salve, Repolar Pharmaceuticals Oy, Espoo, Finland).

Honey salve: For medical purposes, honey is generated by bees that have fed on flowers of the Manuka myrtle (Leptospermum scoparium) in New Zealand and southeast Australia. Honey has been shown to possess a wide spectrum of antimicrobial qualities [19]. Several honey products, prepared for wound care, are widely available all over the world. We used refined Manuka honey, prepared for medical use, which carried the CE mark (Activon Tube^® 25 g, Advancis Medical Ltd., Nottingham, England). Patients were instructed to use the resin or honey salves according to manufacturer guidelines. Briefly, the salves were spread directly onto the clean wound (e.g., cleansed with saline), then the wound was covered with a bandage or gauze. When skin damage was widespread or included cavities or fistulae, the salve was spread at a thickness of at least 1 mm onto a gauze or gauze ribbon, which was then used to fill the cavity or fistula channel. Bandages were changed every 1–3 days, depending on the degree of infection and the amount of wound secretion. In cases of profuse wound secretion, which increased the risk of skin breakdown, a liquid barrier film (Cavilon^®, 3M Skin and Wound Care, St. Paul, MN, U.S.A) was applied temporarily to protect wound edges from maceration.

Ethics, registration, and approval: The study protocol was approved by the Ethics Committee of the Kuopio University Hospital (KUH15101075), and it conformed to the guidelines of the Helsinki declaration. This study was registered at ClinicalTrials. gov (identifier: NCT01868412). All patients were verbally informed of the study course, and all provided written informed consent. For ethical reasons, we could not include a control group that did not receive any active home-care treatment. All included patients had marked, severe skin ulcers or wounds that required attention and active treatment.

Results

All available data of ulcer surface size (area; mm²) and follow-up times (days) for all patients in both salve treatment groups are presented in table format in Figures 3-5. Figure 3 shows data for 9 patients with ulcers that fully healed clinically during the 6-month follow-up (patients symbolized with green color). Figure 4 presents data from 13 patients with ulcers that was not recorded to been healed within the 6-month follow-up. In this group, ulcer size data and follow-up times were, however, available from at least 3 follow-up visits enabling the statistical regression analysis; therefore, in this post hoc regression analysis, we could compute the relationship between ulcer size and follow-up time. For 8 patients, the regression analysis showed statistical evidence of a progressive reduction in ulcer size over time (highlighted in yellow, Figure 4). We concluded that these cases were on a successful, appropriate healing course and showed objective evidence of treatment success, even though the complete ulcer healing had not been clinically achieved and recorded at the last follow-up visit. Figure 5 presents data on 17 patients lost to follow-up. These patients had insufficient follow-up data (£2 follow-up visits) for determining objectively and statistically whether the ulcer was on a proper trend of healing. Among these 17 cases lost to sufficient follow-up, reasons for discontinuing the follow-up were not recorded. However, the demographic and background factors were not significantly different between the dropouts and the patients included in the treatment success analyses.

Healing rate

Ulcers were fully healed clinically in 9 patients during the 6-month follow-up in total (Figure 3). An additional 8 patients displayed a statistically objective trend of ulcer healing, according to the linear regression analyses (Figure 4). Thereby, based on the intent-to-treat (ITT) principle, a significant and objective ulcer healing occurred altogether in 9 of 20 patients (45%; 95%CI: 23- 67%) in the resin group and in 8 of 20 patients (40%; 95%CI: 19- 61%) in the honey group (Figures 3 and 4). Alternatively, based on a per-protocol (PP) analysis, after excluding the 17 cases without sufficient follow-up data or postoperative death (one case), we found successful, proper ulcer healing in 9 of 10 (90%; 95%CI: 71-100%) patients and in 8 of 12 (67%; 95%CI: 40-93%) patients in the resin and honey salve groups, respectively (difference not significant). In the entire study population with sufficient followup data (analysis per protocol; PP), healing occurred in 17 of 22 (77%; 95%CI: 60-95%) patients. In one patient in resin group and in two patients in honey group, the ulcer showed a significant trend toward worsening (Figure 4). Moreover, in two patients in honey group, the ulcer did not show any significant trend towards either healing or worsening (Figure 4).

Ulcers due to SSI versus the leg ulcers present already before surgery

In 5 patients, ulcers were initially classified as complicated operation wounds; i.e., SSIs. However, sufficient follow-up data was available for only 3 of these patients. In all three, the ulcer (wound) either healed or showed a trend toward healing (Figures 3 and 4). When the SSI cases were excluded from treatment success calculations, 14 of 19 patients (74%; 95%CI: 54-93%) with ordinary foot ulcers in CLI healed after revascularization, with either resin or honey salve as a postoperative ulcer care.

Life table analyses

The life table analyses indicated the probability of ulcers healing in the resin and honey groups (Figure 6). We found no difference in probability distributions between the two salve groups, when clinical ulcer healing was chosen to indicate the positive favourable event (Log-rank P=0.783). Similarly, no difference was found between the groups when statistical evidence of successful ulcer healing (patients with evidence of ulcer healing by the regression analysis; Figure 4) were also included as favourable events (Log-rank P=0.567).

Healing time

The observed and predicted healing times are shown for all patients with healed or successfully healing ulcers in Figures 3 and 4. In total, the mean healing time was 87 ±48 days (median 80 days), as observed clinically or estimated with the linear regression model. The mean observed healing time (Figure 3) was 61 ±41 days (median 56 days) in patients with the clinically healed ulcer. The mean estimated healing time (liner regression model) was 116 ±38 days (median 108 days) in patients with the statistically significant trend of ulcer to heal (Figure 4).

The healing time in patients with clinically healed ulcers (Figure 3) was not significantly correlated with the initial ulcer size (correlation coefficient r = 0.29; P=0.447). This was also the case (r = 0.33; P=0.436) in patients with the statistical trend of ulcer to heal (Figure 4), when the healing time was predicted by the linear regression model.

Side effects

One patient in the resin group reported signs of local skin irritation at the ulcer site at the end of follow-up. However, an epicutaneous test did not reveal hypersensitivity to coniferous resin (colophony allergy); the final diagnosis was infectious eczema, based on skin patch test and examination by a consulting dermatologist.

Discussion

The present observations show that the treatment of leg ulcers with arterial revascularization for CLI and with a postoperative antiseptic resin or honey salve as a postoperative topical ulcer care agent resulted in healing of leg ulcers in a markedly high number of cases, even in three quarters of patients. Based on PP principles, 17 of 22 (77%; 95%CI: 60-95%) patients with chronic leg ulcer or SSI showed objective evidence of ulcer healing. When the SSI cases were excluded, 14 of 19 patients (74%; 95%CI: 54- 93%) with preoperative chronic foot ulcers in CLI healed after the revascularization, and with either resin or honey salve as the postoperative ulcer care. In these cases, ulcer was clinically healed or showed statistical evidence that it was progressively and significantly decreasing in size. However, a high number of cases were lost from reliable analyses (18 of 40 patients, 45%) due to insufficient follow-up; thus, the healing rate in an ITT analysis remained relatively low; 17 of the initial 40 cases displayed evidence of objective healing (43%; 95%CI: 27-58%).

In the present study, we could not determine how much the vascular surgery and how much the postoperative topical ulcer care with resin or honey salve contributed to treatment success. In our study population, all chronic leg/foot ulcers in CLI were severe ischaemic lesions, which are classified to stage IV CLIs in the Fontane classification and to stage 5-6 CLIs in the Rutherford classification [22]. The ulcers in sever CLI are necrotic, gangrenous lesions. Therefore, they also require attention and active, serious postoperative ulcer care [24,25]. Unfortunately, due to severe nature of the ulcers, it was not ethically accepted to arrange control groups without any active postoperative ulcer care. Antiseptic resin and honey salves were considered acceptable active options; they are safe, even for long-term topical ulcer care at home, and do not induce resistant microbe strains, and are shown to be beneficial in previous trials on various skin ulcers and wounds [17,20,21]. It is, however, conceivable that the postoperative ulcer care with either the resin or honey salve contributed positively to good postoperative ulcer healing. The present 35 patients had a preoperative history of ulcers that had persisted for months or even years, but then, healed within about 3 months after the limb revascularization combined with the topical postoperative antiseptic salve treatment. On its own, peripheral limb rearterialization is, without doubt, an effective surgical therapy in CLI; indeed, it is certainly the basic requirement for good healing results. In most recent clinical trials on ischaemic leg ulcers after rearterialization for CLI, the healing rates (frequency of cases healed) are reported to vary from 36% to 96% in follow-ups of 6-12 months, and the mean rate being around 60% [25-30]. In present study, the healing rate up to 77% within 6 months (PP analysis) is consistent with the literature and corresponded to a good treatment result.

In present study population, the significant reduction in ulcer size over time in linear regression modelling occurred in 5 of the 6 patients with a clinically healed ulcer (Figure 3). Similarly, the regression modelling of data of ulcer size and follow-up time revealed additional 8 ulcer patients considered to be on a similar proper and successful healing course than the patients with the clinically healed ulcer even though the ulcer in these cases was not fully healed (closured) at the last available follow-up visit (Figure 4). The application of statistics and mathematical analyses of changes in ulcer size (surface area) over the available followup time provides an opportunity to improve the objectivity for conclusions about whether an ulcer in question is really healing or not. Thus, the statistical “post hoc” regression analyses of changes in wound/ulcer size could be useful in cases that have not achieved final ulcer closure, for whatever reason, during the follow-up. Hence, the imputations of the “post hoc” analyses with mathematical regression models could increase the reliability and objectivity of conclusions about the ulcer fate in cases that otherwise would be excluded from ITT-type evaluations and “censored” from the life-table analyses. Both resin and honey salves have long histories in wound care; thus, they are plausible biologically active agents. Coniferous natural resin collected from Norway spruce (Picea abies) has been used as a salve mixed with fat or butter for centuries in Nordic countries for treating skin wounds, sores, pressure ulcers, punctured abscesses, suppurating burns, onychomycosis, and paronychia; it has even been used as a chewing gum [17]. The acids of coniferous natural resin are strongly antimicrobial (antiseptic) [17]. They prevent and destroy bacterial biofilms in in vivo studies, and they might possess general anti-inflammatory and cell-regenerative properties; e.g., the capacity to enhance skin epithelialization [17]. In in vitro tests, this strong antiseptic activity is directed against a wide range of bacteria, yeasts, and fungi, including methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci, and even highly drug-resistant “super bacteria”. In clinical trials, the resin salve is shown to enhance the ulcer healing in patients with pressure ulcers, complicated surgical ulcers, or heals paronychia [17]. This effect probably arises from antiseptic mechanisms and/ or from influences on tissue cytokines and tissue growth factors [17].

Honey has also been used for decades in wound care. In general, honey consists of concentrated sugar, which induces high osmolality at the site of application. Together, the low water and high sugar contents restrict microbial growth and might promote wound healing [18,19]. Medical grade honey is generated from Leptospermum species (i.e., Manuka honey). Apart from osmolality, Manuka honey is reported to have woundhealing properties, including anti-infectious, anti-inflammatory, anti-exudative, antioxidant, and wound debriding properties, in addition to its nutritional activities [18,19]. We did not observe any significant major differences in healing rates between resin or honey salve as postoperative ulcer care agent in the present study, either in standard statistical analyses or in life table analyses. Much larger study populations are needed to detect minor differences in efficacy, if any, between the resin and honey salves as postoperative ulcer care in CLI. The rate of dropouts tends to be high, for manifold reasons, in long-lasting clinical follow-ups, particularly in studies on severe diseases among old people at home care. In present investigation this dropout rate was 18 of 40 cases (45%). Our initial power analysis and the estimation of study group sizes were overly optimistic. Therefore, we were limited in the ability to detect significant changes in efficacy, if any, between the two test salves (coniferous resin vs. medical honey).

Finally, a total of 18 cases (including one patient included into the study population but who died before the study start) were lost and excluded from the analyses. However, they might not have markedly altered the general conclusions about the overall treatment results. The patients excluded from were not significantly different regarding the initial ulcer size, demographic characteristics, or clinical observations, compared to the patients included in analyses. The inclusion of two wound types, i.e., ordinary chronic ischaemic ulcers and complicated operation wounds (SSIs), might be another source for interpretation biases, even though both of these ulcer subtypes are intimately linked to CLI. However, when all 5 patients with SSIs were excluded from the analyses, the main conclusions about the favourable treatment success did not change. Another limitation is that the study could not be blinded, due to the specific, readily recognizable features of resin and honey salves. The study neither had a control group without any active postoperative ulcer care. The severity of necrotic ulcers in CLI did not allow to do that.

Innovation

Surgical limb revascularization for CLI in patients with chronic leg ulcer and the postoperative ulcer care with natural resin or honey salve will result in healing of ulcers even in three quarters of cases within 6 months. This can be demonstrated if the calculations also include the “missed” cases in which data of ulcer size are available from three follow-up visits at least. In these cases, the statistical linear regression modelling of the ulcer size (mean area) over the follow-up time can be applied as an objective criterium of whether or not an ulcer in question is at a trend to heal even though the ulcer would not be clinically closured at the last available time point of the follow-up.

Conclusions

Peripheral arterial revascularization for CLI in patients with chronic leg ulcer or SSI, and with postoperative topical ulcer care with natural coniferous resin or medical honey salve, results in ulcer healing in about three quarters of the cases within 6 months. In this treatment strategy, resin and honey salves have equivalent effect as postoperative topical ulcer care agents. There are no major differences in their efficacy as a postoperative ulcer care agent. A linear regression analysis of data of ulcer size over the postoperative follow-up time increases the objectivity of conclusions about whether an ulcer in question is healing properly or not.

References

1. Park H, Copeland C, Henry S, Barbul A (2010) Complex wounds and their management. Surg Clin North Am 90: 1181-1194.
2. Harding KG, Morris HL, Patel GK (2002) Science, medicine and the future: healing chronic wounds. BMJ 324: 160-163.
3. Clair D, Shah S, Weber J (2012) Current state of diagnosis and management of critical limb ischemia. Curr Cardiol Rep 14: 160-170.
4. Turtiainen J, Saimanen EI, Mäkinen KT, Antti I Nykänen, Maarit A Venermo, et al. (2012) Effect of triclosan-coated sutures on the incidence of surgical wound infection after lower limb revascularization surgery: a randomized controlled trial. World J Surg 36: 2528-2534.
5. Anderson DJ (2011) Surgical site infections. Infect Dis Clin North Am 25: 135-153.
6. Kirby JP, Mazuski JE (2009) Prevention of surgical site infection. Surg Clin North Am 89: 365-389.
7. Pounds LL, Montes-Walters M, Mayhall CG, Pamela S Falk, Ellen Sanderson, et al. (2005) A changing pattern of infection after major vascular reconstructions. Vasc Endovascular Surg 39: 511-517.
8. Damrauer SM, Gaffey, AC, DeBord Smith A, Ronald M Fairman, Louis L Nguyen, et al. (2015) Comparison of risk factors for length of stay and readmission following lower extremity bypass surgery. J Vasc Surg 62: 1192-1200.
9. Zhang JQ, Curran T, McCallum JC, Li Wang, Mark C Wyers, et al. (2014) Risk factors for readmission after lower extremity bypass in the American College of Surgeons National Surgery Quality Improvement Program. J Vasc Surg 59: 1331-1339.
10. Kawarada O, Fujihara M, Higashimori A, et al. (2012) Predictors of adverse clinical outcomes after successful infrapopliteal intervention. Catheter Cardiovasc Interv 80: 861-871.
11. Richet HM, Chidiac C, Prat A, M David, M Maccario, et al. (1991) Analysis of risk factors for surgical wound infections following vascular surgery. Am J Med 9: 170S-172S.
12. Pejkić S, Dragaš M, Ilić N, Igor Končar, Dragan Opačić, et al. (2014) Incidence and relevance of groin incisional complications after aortobifemoral bypass grafting. Ann Vasc Surg 28: 1832-1839.
13. Kobayashi N, Hirano K, Nakano M, Toshiya Muramatsu, Reiko Tsukahara, et al. (2015) Development and validation of a new scoring system to predict wound healing after endovascular therapy in critical limb ischemia with tissue loss. J Endovasc Ther 22: 48-56.
14. Brölmann FE, Ubbink DT, Nelson EA, K Munte, CMAM van der Horst, et al. (2012) Evidence-based decisions for local and systemic wound Br J Surg 99: 1172-1183.
15. Ubbink DT, Brölmann F, Go PM, Vermeulen H (2015) Evidence-Based Care of Acute Wounds: A Perspective. Adv Wound Care 4: 286-294.
16. Forster R, Pagnamenta F (2015) Dressings and topical agents for arterial leg ulcers. Cochrane Database Syst Rev 6: CD001836.
17. Sipponen A (2013) Coniferous resin salve, ancient and effective treatment for chronic wounds – laboratory and clinical studies. Academic thesis. Helsinki University 2013.
18. Biglari B, Moghaddam A, Santos K, Gisela Blaser, Axel Büchler, et al. (2013) Multicentre prospective observational study on professional wound care using honey (Medihoney™). Int Wound J 10: 252-259.
19. Al-Waili NS, Salom K, Butler G, Al Ghamdi AA (2011) Honey and microbial infections: a review supporting the use of honey for microbial control. J Med Food 14: 1079-1096.
20. Jull AB, Cullum N, Dumville JC, Maggie J Westby, Sohan Deshpande, et al. (2015) Honey as a topical treatment for wounds. Cochrane Database Syst Rev 6: CD005083.
21. Dumville JC, Lipsky BA, Hoey C, Mario Cruciani, Marta Fiscon, et al. (2017) Topical antimicrobial agents for treating foot ulcers in people with diabetes. Cochrane Database of Systematic Reviews 6: CD011038.
22. Hardman RL, Jazaeri O, Yi J, Smith, Rajan Gupta (2014) Overview of Classification Systems in Peripheral Artery Disease. Semin Intervent Radiol 31: 378-388.
23. Sipponen A, Kuokkanen O, Tiihonen R, Harri Kauppinen, Janne J Jokinen (2012) Natural coniferous resin salve used to treat complicated surgical wounds: pilot clinical trial on healing and costs. Int J Dermatol 51: 726-732.
24. Varela C, Acín F, de Haro J, Silvia Bleda, Leticia Esparza, et al. (2010) The role of foot collateral vessels on ulcer healing and limb salvage after successful endovascular and surgical distal procedures according to an angiosome model. Vasc Endovascular Surg 44: 654-660.
25. Söderström M, Albäck A, Biancari F, Kimmo Lappalainen, Mauri Lepäntalo, et al. (2013) Angiosome-targeted infrapopliteal endovascular revascularization for treatment of diabetic foot ulcers. J Vasc Surg 57: 427-435.
26. Georgopoulos S, Kouvelos GN, Koutsoumpelis A, C Bakoyiannis, M Lymperi, C Klonaris, et al. (2013) The effect of revascularization procedures on healing of mixed arterial and venous leg ulcers. Int Angiol 32: 368-374.
27. Nakama T, Watanabe N, Haraguchi T, Hiroshi Sakamoto, Daisuke Kamoi, et al. (2017) Clinical outcomes of pedal artery angioplasty for patients with ischemic wounds: Results from the multicenter RENDEZVOUS registry. JACC Cardiovasc Interv 10: 79-90.
28. Kabra A, Suresh KR, Vivekanand V, Motukuru Vishnu, Raj Sumanth, et al. (2013) Outcomes of angiosome and non-angiosome targeted revascularization in critical lower limb ischemia. J Vasc Surg 57: 44-49.
29. Darling JD, McCallum JC, Soden PA, Lindsey Korepta, Raul J Guzman, et al. (2017) Results for primary bypass versus primary angioplasty/stent for lower extremity chronic limb-threatening ischemia. J Vasc Surg 66: 466-475.
30. Sadaghianloo N, Jean-Baptiste E, Declemy S, Aurélien Mousnier, Sophie Brizzi, et al. (2013) Percutaneous angioplasty of long tibial occlusions in critical limb ischemia. Ann Vasc Surg 27: 894-903.