The use of medicinal plants in venous ulcers: a systematic review with meta‐analysis

ABSTRACT

The aim of this study was to systematically review the use of medicinal plants for the treatment of varicose ulcers. The databases used in the search were: Medline/Pubmed, Scopus, Cinhal, Lilacs and Web of Science. The selection process was divided into two phases: the reading of titles and abstracts and the full reading of selected articles. The item description was compared with the checklist of the Consolidated Standards of Reporting Trials. The initial search produced 3505 articles and seven were selected for inclusion in the systematic review. Of the included studies, 7 (100%) evaluated the reduction of the ulcer area, 4 (57·14%) evaluated reepithelisation, 2 (28·57%) evaluated bacterial flora and 1 (14·28%) evaluated the oxygen pressure and percutaneous carbon dioxide. The level of evidence rating indicated that five studies (71·42%) were rated at level 2 and two (28·57%) were rated at level 3. The quality assessment was performed using the Jadad scale, which is prevalent in the literature. The quality score of the Jadad questionnaire ranges from 0 to 5; here, the studies analysed had an average of 2·5. A meta‐analysis was performed on two studies that analysed the effects of Mimosa tenuiflora hydrogel in the treatment of venous ulcer and included 42 patients with a mean age of 60·5 years and a mean duration of treatment of 10·5 weeks. Heterogeneity was assessed using I²; we obtained a high value of 84%. We concluded that, despite the efficacy of the incorporation of Ageratina pichinchensis into the gel, the hydrogel that incorporated M. tenuiflora appeared to be a promising candidate for the management of venous ulcers.

Introduction

In humans, problems related to blood circulation, such as chronic venous ulcers, are common. This disease is associated with the complication of chronic venous insufficiency, which is the leading cause of lower limb ulcers. Venous ulcers are caused by the malfunction of the venous valves, which affects the oxygen and nutrient supply to tissues and is conducive to the appearance of cutaneous lesions 1. Most often, individuals between 60 and 80 years old are affected; however, the majority of injuries start before 60 years 2. A comparison of the data from a number of Western countries indicated the epidemiological prevalence varied from 0·7% to 2·7% 2, 3

The treatment cost of the disease is generally high and results in significant social and economic impact owing to the lengthy process of the treatment and cicatrisation 2, 3, 4. The main treatments reported for venous leg ulcers are compression therapy (single layer, highly elastic or multilayer) and the traditional Unna boot 4, 5, 6.

The search for new therapeutic options for the treatment of venous ulcers has been studied; particular emphasis has been placed on therapies using medicinal plants 7. According to survey data, the knowledge of medicinal plants that has been passed down through for generations indicates that a population, generally in a rural area, may possess extensive knowledge on the subject, despite the absence of pharmacological studies of many of the plants used 8, 9.

The use of medicinal plants is often chosen because of their low cost or as an alternative when conventional treatment is ineffective 10. Considering the prevalent and new treatment options for varicose ulcers, this study aimed to conduct a systematic review of the use of medicinal plants for the treatment of varicose ulcers.

Methodology

Research strategy

The research evaluated all articles published before the end of August 2016, in Portuguese, English and Spanish, in the following databases: Cinhal, Lilacs, Medline/Pubmed, Scopus and Web of Science. The mesh terms in English were: Plants, Medicinal; Wound Healing and Varicose Ulcer. In Portuguese, the search terms were: Plantas Medicinais; Úlcera Varicosa; Cicatrização and Boolean operators were included between the terms. The individual search was conducted by two reviewers (ALF and CA) and if in doubt, a third reviewer was consulted.

Inclusion and exclusion criteria

Clinical trials were included if they had evaluated the healing of venous ulcers with the topical use of herbal medicines. Studies that evaluated ulcer healing with the use of herbal medicines in animals 11, ulcers with other etiologies 12, 13, cost‐benefit 14 and reporting cases were excluded 15.

Studies that assessed healing through herbal topical phytotherapeutics were also included, but assessments of only oral phytotherapeutics were excluded 16.

Data extraction

The data were extracted and a spreadsheet was populated with the relevant data items, such as: title, database, place of publication, author, year, objectives, study type, processing of data by intention to treat or via protocol, age range of patients, collected variables and measurement methods of healing and completion of the study. The data were independently extracted by two researchers (ALF and CA) and the results were compared.

The Consolidated Standards of Reporting Trials (CONSORT) is a consensus method for the description of clinical trials; this checklist was used to determine a score between 0 and 25 points 17.

The risk of the bias was evaluated by the Cochrane Collaboration tool for the assessment of bias in randomised clinical trials, which is available in RevMan 18.

Statistical analysis

With regard to the meta‐analysis, only two studies were included because the studies used the same kind of medicinal plant and formulation applied in the treatment of venous ulcers.

A direct meta‐analysis was performed; the data were collected by the reviewers and transferred to the RevMan, which is available for free from the Cochrane Collaboration.

To determine the efficacy of dichotomous variables, we used the relative risk as a measure of the effect of the intervention and assessed the odds ratio for achieving a cure. On the assumption of heterogeneity among the selected studies, which arose from the different treatment periods, we used the random effects model, which distributed the weight assigned to studies more evenly. The chosen random method was the inverse of the variance and confidence interval of 95% 19.

Heterogeneity among studies was analysed by the statistical parameter I², which is commonly used because it is easy to interpret. The I² value is between 0% and 100%: the heterogeneity is considered low when I² is less than or equal to 25%; I² values between 25% and 50% reflect moderate heterogeneity between studies; and a value greater than 50% is considered to represent high heterogeneity 19.

Sensitivity analyses were performed to detect possible sources of heterogeneity in the meta‐analysis. First, the statistical models and methods employed were altered to the fixed effects model and the occurrence of any substantial change in the value of I² was observed. Subsequently, we evaluated the impact of the hypothetical withdrawal of each study on the overall result of the meta‐analysis.

Results

The results of the individual search made by two researchers (ALF and CA) are demonstrated sequentially on the fluxogram (Figure 1).

Figure 1.

Selection of the articles.

From the total 3505 items, 110 articles were excluded owing to repetition, 3385 abstracts were excluded during the phase of title and abstract reading, 21 articles were eliminated during the reading phase of the articles in their entirety as they did not meet the inclusion criteria.

Characterisation of studies

The study comprised seven primary articles. Two studies (28·57%) addressed topical treatments in clinical trials. Among the topical therapy studies, 2 (28·57%) evaluated treatments with hydrogel, 2 (28·57%) evaluated treatment with cream, 1 (14·28%) evaluated treatment with gel, 1 (14·28%) evaluated treatment with plant biomembrane and 1 (14·28%) evaluated treatment with capsule powder.

The species used in the studies were Mimosa tenuiflora (two studies), Allii bulbus, Hypericum perforatum and Calendula officinalis (one study), Pinus pinaster (one study), C. officinalis, Symphytum officinalis, Achiléa millefolium and Salvia officinalis (one study), Hevea brasiliensis (one study) and Ageratina pichinchensis (one study).

With regard to the design of the studies, five studies (71·42%) were randomised (of which one was a pilot study) and two (28·57%) were not randomised (of which one was a pilot study).

With regard to the analysed results, 7 studies (100%) evaluated the reduction in ulcer area, 4 (57·14%) evaluated reepithelialisation, 2 (28·57%) evaluated the bacterial flora and 1 (14·28%) analysed the pressure of percutaneous oxygen and carbon dioxide.

According to the classification of evidence levels, five studies (71·42%) were rated at level 2 and two (28·57%) were rated at level 3 20.

To evaluate the description of the items entered in the review, an average of 17·92 (71·68%) points was obtained; the study of Belcaro et al. had the lowest score and the study of Romero‐Cerecero et al. had the highest score (22·5 points). The minimum possible score was 0 and the maximum was 25 points 17, 21, 22.

The risk of bias was assessed by the RevMan 18; the risk of bias was unclear or moderate in most fields and there was high risk in the fields of randomisation and allocation of patients. The graph obtained is shown in Figure 2.

Figure 2.

Analysis risk of bias.

The quality assessment study was performed by the Jadad scale, which is widely used in the literature. In the Jadad questionnaire, the quality score ranges from 0 to 5; the studies analysed had an average of 2·5 23.

The main characteristics of the studies are summarised in Table 1.

Table 1.

The main characteristics of the studies

Author/year	Average age/M and W	Length	Measurement of cicatrisation	Average rate of reepithelisation (%)	Used plant	PP/ITT	Study design	Pharmaceutical form
Kundakovc et al., 2012 33	72·6/10 M and 15 W	7 weeks	Planimetric in photography	99·1	Allii bulbus, Hypericum perforatum, Calendula officinalis	PP	Non‐randomised pilot clinical trial	Cream
Belcaro et al., 2005 22	56·6/10 M and 8 W	6 weeks	Ruler	100	Pinus pinaster	PP	Clinical trial	Powder capsule
Binic et al., 2010 26	66·8/14 M and 18 W	7 weeks	Ruler	11·76	Calendula officinalis, Symphytum officinalis, Achilea millefolium, Salvia officinalis	ITT	Randomised and controlled clinical trial.	Cream
Frade et al., 2012 5	62·8/5 M and 16 W	120 days	Planimetric in photography (Image J)	42	Hevea brasiliensis	ITT	Randomised clinical trial	Biomembrane plant
Lammoglia‐Ordiales et al., 2012 24	60/13 M and 19 W	8 weeks	Planimetric in photography (Image J)	22	Mimosa tenuiflora	PP	Randomised double‐blind clinical trial.	Hydrogel
Romero‐Cerecero O et al., 2013 30	61/10 M and 24 W	10 months	Planimetric in photography	100	Ageratina pichinchensis	ITT	Randomised double‐blind clinical trial.	Gel
Rivera‐Arce et al., 2007 7	61/40 M and W	13 weeks	Planimetric in photography	100	Mimosa tenuiflora	PP	Randomised, controlled double‐blind clinical trial.	Hydrogel

ITT, intention‐to‐treat analysis; M, man; PP, analysis by protocol; W, woman.

Meta‐analysis

The meta‐analysis was performed on two studies that analysed the effects of M. tenuiflora hydrogel in the treatment of venous ulcer and included 42 patients with a mean age of 60·5 years and a mean treatment duration of 10·5 weeks 7, 24. Heterogeneity was assessed using I²; a high rating of 84% was obtained. The Forest Chart is presented below and the main characteristics of the two studies are shown in Figure 3.

Figure 3.

Forest chart.

Discussion

The application of compression therapy to increase venous ulcer healing is already recommended by the guidelines for clinical practice of the Vascular Surgery Society 25. Topical therapy has not yet been established; a bandage that absorbs ulcer exudate has been suggested, but the bandage composition has not been set. The incorporation of natural products into these dressings has been shown to offer potential anti‐inflammatory, antibacterial and antioxidant activities, which greatly assist the healing of venous ulcers 25, 26, 27, 28.

Among the included studies, the clinical trial described by Belcaro et al. observed 100% healing of ulcers within 6 weeks using P. pinaster, which was attributed to its antioxidant activity as the use of antioxidant substances has been related to the reduced formation of free radicals and lower tissue damage in hypoxic situations 21, 28, 29.

Using A. pichinchensis, Romero‐Cerecero et al. also observed 100% healing of venous ulcers within 10 months; the healing activity was attributed to the acute and chronic anti‐inflammatory activities and the induction of cell proliferation caused by the A. pichinchensis 22, 30.

Similar results of 100% ulcer healing in 13 months using M. tenuiflora were also observed by Rivera‐Arce et al. 7 The function of tannins in the healing process is not fully clear; however, tannins are known to possess antimicrobial properties that occur through different mechanisms, including inhibition of the enzyme, a reduction in oxidative phosphorylation and iron deprivation 31.

Despite having used the same plant (M. tenuiflora), the study of Lammoglia‐Ordiales et al., did not report the same results, but instead found 22% ulcer healing within 8 weeks. This may be because Rivera‐Arce et al. did not determine the size of the initial ulcer, which may have influenced the cicatrisation rate. Although the study period of Lammoglia‐Ordiales et al. was shorter, it had already been established by Gelfand et al., that a monitoring period of 4 or 8 weeks was sufficient to predict the healing progress at 2 weeks 7, 24, 32.

Another study by Kundakovic et al. found 99·1% healing in 7 weeks, using A. bulbus, H. perforatum and C. officinalis. The antiseptic and bacteriostatic activity can be attributed to the activity of the compounds in A. bulbus, called allicin. H. perforatum and C. officinalis are traditionally used in wounds and burns and show healing activities and properties when separate or together 33, 34, 35, 36.

The study of Frade et al. found 42% ulcer healing in 3 months using H. brasiliensis, which promotes the neoangiogenesis of newly formed tissue, as this healing did not occur with neoangiogenesis and the cells were dependent on blood supply for their maintenance 5, 37.

The study of Binic et al. demonstrated a healing rate of 11·76% in 7 weeks, the lowest rate of the included studies, and used C. officinalis, S. officinalis, A. millefolium and S. officinalis. Although C. officinalis possesses anti‐inflammatory activity in vivo, there is a lack of clinical evidence to corroborate this statement, as well as bacterial and healing effects that may have led to low cicatrisation. The S. officinalis plant possesses anti‐inflammatory and anti‐exudative activity, which may confer the potential of cicatrisation. Furthermore, A. millefolium possesses healing and anti‐hemorrhagic activities. Another plant also used in the formulation, S. officinalis, has antibacterial, antioxidant and anti‐inflammatory activities. The use of these plants is relevant with regard to their individual activity, but does not guarantee their effectiveness if used together 27, 38, 39, 40, 41.

The calculation of the samples was not reported in three studies, furthermore, two were pilot trials and another had a small sample size. The study by Frade et al. did not demonstrate significant differences, but was affected by the small number of patients. Moreover, seven of the patients (50%) experienced a recurrence of ulcers compared with the control group, in which two studies had demonstrated recurrence (28%) 5, 27, 33, 42.

The evaluation of the quality of the studies was performed by the Jadad scale; the average value of 1·4 demonstrated that the articles had a low methodological quality and/or did not clearly describe the methodological procedures.

The assessment of risk of bias was unclear or high risk, which occurred in most cases when suggestions for the wording of articles were not in line with the CONSORT guidelines, or from methodological flaws in their studies 17, 23.

This meta‐analysis, which was performed on two studies that investigated the treatment of venous ulcer with hydrogel embedded with M. tenuiflora, presented heterogeneous results. This observation could be explained by the high rate of premature discontinuation in the control and treatment groups 24. The heterogeneity of the results impaired the statement that hydrogel incorporating M. tenuiflora was an effective treatment for venous ulcers. However, when we considered the fixed effect and not the random effects model, the results were more promising.

Therefore, we concluded that despite the efficacy of A. pichinchensis incorporated into the gel, the hydrogel that incorporated M. tenuiflora appeared to be a promising candidate for the management of venous ulcers.

Freitas A. L., Santos C. A., Souza C. A. S., Nunes M. A. P., Antoniolli Â. R., da Silva W. B., da Silva F. A.. The use of medicinal plants in venous ulcers: a systematic review with meta‐analysis. Int Wound J 2017; doi: 10.1111/iwj.12751