Protocol development

This review will be performed following guidelines from the Preferred Reporting Items for Systematic Reviews and Meta-analysis for Protocols (PRISMA-P) (Additional file 1) [29] and the protocol guidelines for Systematic review and meta-analysis for animal intervention studies [30]. The final review will be conducted and reported with reference to PRISMA 2020 statement [29] and published in a peer-reviewed journal. This protocol has been submitted for registration in PROPSERO (https://www.crd.york.ac.uk/prospero/) (Application number 330330).

Review question

Is there a difference between the percent wound closure of induced wounds in animal wound models treated with nano encapsulated herbal extracts and that of other alternatives used? The PICOS model for the review question is shown in Table 1.

Table 1 PICOS model for the review questionArticle search

Articles on the efficacy of herb-loaded nanomaterials on in vivo wound healing published from 2000 (there is no available data before this year) to date will be searched from the following electronic databases: Web of Science, MEDLINE Ovid, Pub Med, EMBASE, and Google Scholar.

The following grey literature sources will also be searched: The Agency for Healthcare Research and Quality (AHRQ), Bielefeld Academic Search Engine (BASE), National Institute for Health and Care Excellence (NICE), Networked Digital Library of Theses and Dissertations (NDLTD), Open Grey, ProQuest Dissertations & Theses, National Library of Medicine (NLM), and World Health Organization Institutional Repository for Information Sharing (WHO IRIS).

Search strategy

The search will be conducted by a qualified librarian, Alison A. Kinengyere, as an information retrieval specialist. An electronic search will be conducted with the following search terms and their Medical subject heading (MesH) in abstract, keyword, and text.

Animal (rats or mice or rabbits or pigs) to represent the population.

Plant (herb or crude extract or plant extract or herbal medicine or phytochemical or leaf extract or essential oil) as the main intervention.

Nano (nanoparticles or nano-composites or nano-capsules or nano-spheres or nano-medicine or nano-liposomes or nanofibers or nanomaterials or nano-gels or dendrimers or cyclodextrins). To capture aspects of the intervention.

Wound healing, (wound or wound model or excisional wound or incision wound or murine wound or dead space or burn wound or infected wound or diabetic wound or splinted wound or cutaneous wound) to capture aspects of the outcome.

The search terms will be combined using Boolean logic “OR” for related terms and “AND” for terms from different elements of PICOS. Truncation and wildcards will be added to terms where applicable.

A stepwise serial search pilot example from PubMed is provided (See Additional file 2). Once papers are included, the reference lists of those retrieved papers will be checked for additional eligible studies. In addition, authors will be contacted to get the missing information. The retrieved articles will be imported using endnote software and duplicates removed.

Selection criteria

Articles will be screened for inclusion and exclusion according to the following criteria:

Study design

Inclusion: experimental laboratory design setting using animal wound models. Studies that compare with control group.

Exclusion: studies without controls.

Population

Inclusion: Animal (rats, mice, rabbits and pigs/porcine) with laboratory induced wounds will be included.

Exclusion: wound healing studies in vitro, ex vivo, in silico, or in humans.

Intervention

Inclusion: Studies that evaluate wound healing where herbal crude extract or phytochemical or plant oil is encapsulated inside the nanomaterial.

Exclusion: Studies involving plant synthesized metallic nanomaterials. Studies on wound healing using any herbal crude extracts or phytochemicals or oils that have not been encapsulated into a nanocarrier will not be considered.

Outcome measure Others

Inclusion: only original peer-reviewed papers in the English language published from the year 2000 to date.

Exclusion: case reports, conference papers, reviews, editorials, unpublished work, and letters to the editor.

Screening and selection of articles

The literature will be retrieved from the listed databases using the mentioned search strategies and the findings reported using the PRISMA 2020 statement [29]. Endnote reference management software will be used to manage citations, bibliographies, export and import citations, as well as de-duplicate retrieved studies. We will contact the authors of retrieved studies for missing data or additional data, where required.

Groups of included and excluded studies will be created in endnote as follows:

Group 1 will be named included studies, where all studies that meet the inclusion criteria will be put.

Group 2 will be named excluded studies, where all studies that do not meet the inclusion criteria will be put.

Group 3 will be named unresolved, where studies awaiting a tiebreaker to resolve will be put.

The three folders will be used during Title/Abstract screening. After Title/Abstract screening, a fourth folder will be created and named included final, where studies that meet the inclusion criteria after full-text screening will be put.

Two review team members will independently screen titles and abstracts of retrieved studies to identify studies that potentially meet the inclusion criteria.

Before obtaining the full text of retrieved studies or literature, the results of this screening process will be compared and discussed to reach a consensus concerning the studies to be obtained in full-text format.

Two review team members will independently review the full text of studies whose titles and abstracts have been screened eligible using the inclusion criteria. Any disagreements that may arise among the review members will be resolved by consensus or with reference to a third team member as appropriate.

Data extraction

We will develop standardized data extraction spreadsheet forms in Excel. Three team members will independently extract data from included studies onto the form. Data extraction forms will be used in extracting data from the selected studies with the help of the ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines [31]. The abstracted data will be kept in Excel 2013 file. The form will be piloted on five studies to check for expected review outcomes, and basing on the findings, adjustments will be made. The level of agreement between reviewers will be determined by the Kappa statistic (> 0.75 for excellent, 0.40–0.75 moderate, and < 0.40 for poor) [32]. Any further disagreements will be referred to a tiebreaker. Extracted data will mainly be based on the key PICO components addressing our research question. Information that will be extracted from studies includes:

Study ID: country, author, year of publication, publication status, journal of publication.

Population: animal (type of animal, strain, species, gender, weight, housing, and feeding conditions).

Condition of the wound: wound model, wound type, location, wound size, severity, infected or non-infected.

Intervention: formulation, concentration, dose, administration route, administration frequency, duration, type and formulation of the nanocarrier.

Comparator: number of controls, type of controls.

Outcome measure: wound contraction rate (%wound closure), tensile strength, histopathology results, the number of days to complete healing, whether healing left a scar or not, immunohistochemical analysis.

Study design: sample size calculation, sample size, number of experimental units, number of animals per group, sampling methods.

Randomization methods: whether blinding was done and at what stages it was done.

We will also find out whether there are animals that were excluded from the analysis (dropouts) and reasons for their exclusion.

In incidences where the study used multiple interventions, only the data relating to our research question will be extracted. Data will be extracted from graphs, text, and tables. Graphical data will be extracted using digital screen ruler [33]. Where necessary, we shall contact the authors for missing information.

Outcome measure

Our primary outcome will be the wound healing rate expressed as the mean percentage proportion of a completely healed wound (wound closure) defined as %Wound closure = \(\frac\)×100 (where; \(A0\) is the area of the wound at day 0, and \(An\) is the area of the wound on the nth day after wound induction).

Secondary outcome measures will include the following: the rate of re-epithelization, the time to complete wound healing, wound tensile strength, histopathology (granulation tissue formation, re-epithelization, neovascularization, fibroblast proliferation, collagen synthesis, presence of hair follicles, inflammatory response), hydroxyproline, hexosamine and hexuronic content, and immunohistochemical analysis.

Quality assessment

We anticipate that all included studies will use experimental laboratory design using animal models. Therefore, we will use the Systematic Review Centre for Laboratory Animal Experimentation (SYRCLE) risk of bias tool for animal studies [34] to assess the data quality and risk of bias. The tool was derived from the Cochrane Collaboration Risk of Bias Tool. It is adapted to assess methodological quality and features of bias such as selection bias, performance bias, attrition bias, and reporting bias [34]. Two review members will assess the risk of bias. Any disagreements will be resolved through discussion with the involvement of a third party. Two team members will individually assess the quality of evidence across all the included studies for the outcome of interest. They will use the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach [35] for summary assessment of the certainty of evidence. Any discrepancies will be resolved by discussion and consensus or referral to a third member if necessary.

Data analysis

A quantitative and narrative synthesis of the studies included will be presented using a descriptive data table. We will report the experimental outcomes of the studies. These will include study setting, study design, population, intervention, outcome measures, complications or adverse effects, and any other suitable findings of each study. In this narrative evaluation, we shall comment on whether the efficacy of herb-loaded nanomaterials on wound healing appears to vary according to the intervention subgroups. If the findings are suitable for meta-analysis, we shall first take a heterogeneity analysis related to study design, population, interventions, comparators, and outcomes.

Dichotomous data will be analyzed using risk ratios, while continuous data will be analyzed using mean differences or standard mean differences. Statistical heterogeneity will be assessed using chi square and quantified using the I2 statistic. The threshold for I2 value will be interpreted as follows: 0–25% for very low heterogeneity, 25–50% for low heterogeneity, 50–75% moderate heterogeneity, and above 75% for high heterogeneity [36]. We shall use random-effects model because of the nature and differences in animal studies [33]. Subgroup analyses will be performed to further interrogate primary and/or secondary outcomes based on the type of nanocarrier (nanoparticle, nanoliposome, nanocomposite, nanofiber, nanoemulsion, hydrogel). To also ascertain whether either of the main intervention types (i.e., crude extract, isolated phytochemical, essential oil) is superior, we will present the data according to the herbal intervention used if sufficient data are available. If the findings of the meta-analysis are robust, we shall perform a sensitivity analysis [33]. The analysis will done in STATA 15 and CAMARADES data-manager. Publication bias will be assessed using funnel plots as well as trim and fill if necessary.