Application of genus Cassia in the treatment of Constipation: A systematic review

Purpose: Role of genus cassia in the treatment of Constipation Methods: Methodological analysis, systematic review, and meta-analysis of identified studies using RevMan Result and Discussion: Cassia fistula was partially effected in treating constipation however there is a need for improvement in the protocol of studies to reduce biases. These results were only limited to one species so it cannot be generalized among all species of Cassia. Conclusion: Cassia fistula is partially effective in reducing the pain and consistency of stool during constipation among children.


Introduction
Constipation is a clinical disorder attributed to ineffectual colonic impulsion and/or increased resistance to the proliferation of colonic matters 1 .Approximately 20% of the world population suffers from chronic constipation 2 .It is one of the most common pediatric problems 3 .It was found to be the second most stated disorder in the field of pediatric gastroenterology.Treatment costs for children with constipation will be around three times higher than children without constipation in the United States 4 .African American children, particularly girls, are greatly affected by constipation, which has been associated with poor hygiene conditions 5 .

Literature screening
Identification of articles were performed at level 1 using the search strategy as mentioned in Table 1.Duplicate articles, irrelevant articles such as polyherbal formulation, review articles, or any article other than Cassia or Senna were removed at level 2. Only four 4 articles were identified as being relevant.One record was excluded due to not being a full text article.Abstracts were being reviewed for the following inclusion and exclusion criteria at level 4 and one article was removed for not meeting the eligibility criteria i.e.Randomized, clinical trial on Constipation, full-length open access articles, Pediatric Functional Constipation (age range: 2-15 years).

Eligibility criteria
Types of studies.The author has selected studies of randomized open label, prospective, controlled, parallel-group clinical trial for meta-analysis.Baseline characteristics of randomized trials  of studies included on pediatric functional constipation are presented in Table 4. Characteristics of the studies included are mentioned in Table 5.

Types of participants.
The author included studies involving patients (aged 2-15 years) with Functional constipation.The diagnosis of Functional constipation was according to according to the Rome III criteria 26 .Inclusion and exclusion criteria were based on Study design, participants, intervention, outcome (SPIO) criteria and indicated in Table 2.

Types of interventions.
Included studies were focused on the role Cassia in the treatment of Functional constipation.Unfortunately, there were only two studies identified.

Types of outcomes.
Eligible studies included consisted of the following outcomes: improvement in the episodes of fecal incontinence per week, improvement in the episodes of retentive posturing per week, improvement in the average of severity of pain of defecation (by VAS), improvement in defecation frequency per week, patient's drug compliance and improvement in the average of consistency of stool defecated (by VAS).

Methodological quality assessment (MQ)
Methodological quality assessment was made on the basis of following criteria.1) Aims and Hypothesis clearly defined, adequate sample representation, patient care quality, ethical approval protocol, outcomes assessment, validity and reliability of outcome measure, attempt to blind researcher, follow-up, appropriate statistical analysis and missing data reported.Ten item defined evaluation of methodological quality (MQ) is presented in Figure 2. Risk of Bias were assessed using Cochrane collaboration's tool on the basis of the following criteria such as  selection bias, performance bias, attrition bias, reporting bias and miscellaneous.Cochrane Collaboration's tool for assessing the risk of bias was used and the results are presented in Table 6 27 .

Data extraction
The following data were extracted according to study characteristics (e.g., first author, year of publication, search dates, and number of included studies), patient characteristics (functional constipated children, aged between 2-15), sample size, study type (e.g., Randomized open label, prospective, controlled, parallel-group clinical trial study), randomization methods (e.g., "systematic randomization and simple randomization") and outcome measures/variables (e.g., improvement in defecation frequency per week, improvement in the episodes of fecal incontinence per week, improvement in the episodes of

Length of each contact with the participant/ caregivers
Clinical efficacy and tolerance were assessed using weekly sheets, parents completed every night.They were given three sheets (included seven questions in seven columns) to complete them daily for 3 weeks.
During the study, we had regular phone calls with the parents to check the probable complications, treatment (taking the prescribed drugs) and data filling process.If there were any serious questions or problems, we visited the child.At the end of 4 weeks of treatment, the children were visited and the filled out forms were taken and evaluated.

Blinding of Experiment
The investigators, the children and their parents were aware of the study group assignment.
Fortunately, due to the developing socioeconomic conditions of the people in these regions in recent years, we were able to keep in touch with all the patients during the study period by phone call.

Statistical analysis
Meta-analysis was conducted using the Review Manager (RevMan) 5.3 software 28 .The summary measures were reported as odds ratios (ORs) or as a standard mean difference (SMD) with 95% confidence intervals (CI).The presence of heterogeneity among trials was assessed using the Chi-square test, and the extent of the inconsistency was measured by I2 statistics.Output file from RevMan is available as Underlying data 29 .

Results and discussion
Both of the selected studies were not blinded during intervention and outcome assessment that will result in performance bias and detection bias respectively.These biases occur where the investigators know about the participant's treatment group.Performance bias can also refer to the fact that participants can change their responses or behaviour if they know which group they are allocated in.Blinding of outcome assessment may decrease the risk of the investigator or participant being aware of the treatment that a patient is receiving.If the participants and the caregivers are aware of the intervention and outcome that may affect the behavior of the participants, these behavioral changes may affect the performance of the treatment.Clinical trials on adults were also excluded, such as a randomized clinical trial of a phytotherapeutic compound containing Pimpinella anisum, Foeniculum vulgare, Sambucus nigra, and Cassia augustifolia for chronic constipation 9 .Results of both included studies were non-significant when comparing their baseline characteristics of pediatric functional constipation as presented in Table 5.
During analysis of study characteristics, it was found that both of studies demonstrated Cassia fistula is helping to treat constipation among the children as shown in Table 3, but there is a risk of bias according to Cochrane Collaboration's tool (Table 6).Moreover, both studies found were from one country (Iran).
During a methodological assessment, many flaws were identified   such as inadequate patient care, attempt to blind the researcher and missing data (Figure 2).During meta-analysis, the comparison was made before and after treatment among different variables such as defaecation, fecal incontinence, retentive posturing, the severity of pain, and consistency of stool.All the variable (before and after treatment) were found to be symmetrical when plotted on a funnel plot as shown in Figure 4, Figure 6, Figure 8, Figure 10, Figure 12, and Figure 14 respectively.The overall effect for some variables is statistically insignificant (P=0.11,P=0.49, P=0.24) such as fecal incontinence, retentive posturing, and acceptance, tolerance respectively.High heterogeneity was found in two variables i.e severity of pain (90%) and consistency of stool (77%).All the forest plot of defaecation, fecal incontinence, retentive posturing, severity of pain, consistency of stool and acceptance and tolerance are represented in Figure 3, Figure 5, Figure 7, Figure 9, Figure 11 and Figure 13 respectively.

Conclusion
After evaluation of results, it was found that Cassia fistula was not completely effective.It was partly effective in reducing the pain and consistency of stool during constipation.However, these results cannot be generalized among all population.A well designed, expert validated protocol is required in the future.
There is a need to develop an instrument that will be free from bias.Moreover, the results cannot be generalized among all species of Cassia as the studies available are only for one species.
There is a need to isolate identified bioactive compounds from          Are the rationale for, and objectives of, the Systematic Review clearly stated?Yes

Are sufficient details of the methods and analysis provided to allow replication by others? Yes
Is the statistical analysis and its interpretation appropriate?Yes

Are the conclusions drawn adequately supported by the results presented in the review? Yes
If this is a Living Systematic Review, is the 'living' method appropriate and is the search schedule clearly defined and justified?('Living Systematic Review' or a variation of this term should be included in the title.)Yes Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Natural products, Phytochemistry, and Gut motility.

Massimo Bellini
Gastrointestinal Unit, Department of Translational Research and New Technologies in Medicine and Surgery, University of Pisa, Pisa, Italy I've no further particular methodological concerns.My main concern is still the same: I'm really puzzled about the usefulness of a review which takes into consideration only two papers and altogether only 132 patients.

Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes
Are sufficient details of the methods and analysis provided to allow replication by others?Yes

Are the conclusions drawn adequately supported by the results presented in the review? Yes
If this is a Living Systematic Review, is the 'living' method appropriate and is the search schedule clearly defined and justified?('Living Systematic Review' or a variation of this term should be included in the title.)Yes Competing Interests: No competing interests were disclosed.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Version 1
Reviewer Report 28 June 2019 https://doi.org/10.5256/f1000research.19569.r50310 © 2019 Shokraneh F. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Farhad Shokraneh
University of Nottingham, Nottingham, UK The author investigates the effect of Cassia fistula on three outcomes based on two included studies.It seems a valuable research that could potentially direct the pharmaceutical companies toward a new agent to treat the constipation.

Title
The title should be "Cassia fistula for treatment of children and adolescents with constipation: A systematic review of randomized controlled trials".

Authorship
The systematic review requires a process that involves at least two people for screening and data extraction.I suggest the author to find another systematic reviewer to be involved in this work.

Abstract
The abstract is very brief and sentences are incomplete.Abstract should be re-written following the example from similar systematic reviews and following PRISMA for Abstract.

Minor edits
"act an effective as a laxative" should be "act as effective as laxatives".1.
"search data" should be "search date".2.

Methods
The standard search filter for finding RCTs as reported in Cochrane Handbook should be used.

1.
It is advised in Cochrane Handbook searching Embase, MEDLINE/PubMed, and CENTRAL for systematic reviews of RCTs and this review reports searching only MEDLINE/PubMed.

2.
Excluding closed access papers without trying to order them from library or contacting the authors introduces full-text or open access bias.

3.
The structure of the method should change and start with 1: Eligibility Criteria 2: Databases and search strategy 3: Screening process 4: Data extraction process 5: Assessment of risk of bias process 6: Data analysis plan.Currently, the first section of the methods reports info on search and eligibility criteria while there are specific subheadings for these sections.

4.
The rationale behind excluding studies older than 10 years from search date is not clear.5.
The reported number of search results 2207 is not reasonable.A proper search should not find more than 100-200 results for this topic.

6.
The search date has not been reported.7.
Following a published Cochrane review of RCTs in terms of structure and all the items in PRISMA Checklist for full systematic reviews is recommended.

8.
Constipation is missing from search strategy and MeSH terms in the table.9.
The numbers in PRISMA flow diagram does not match the numbers reported in the text and it does not seem to be correct.Following one of the existing published systematic reviews of RCTs from Systematic Reviews journal or Cochrane Database of Systematic Reviews is recommended.

10.
The quality of included studies is being assessed using Cochrane Risk of Bias tool.The role of MQ and using Atlas.ti is not clear in this review.These two are not usually used for SR of RCTs.

11.
Studying the meta-analysis section in the Cochrane Handbook which is freely available online could help revising the statistical analysis.12.The time point for the outcome is not clear.14.
The outcome measure has not been described properly.15.
The complications and adverse effects have not been reported in the review.16.

Results
The sample size reported in the Data Extraction table does not match the numbers in metaanalysis.For example the total for experimental and control group in Review Manager file is 109 for Esmaeilidooki 2016 and 71 for Mozaffarpur 2012 while in the table is 51 and 81 respectively. 1.
The analysis for before treatment is not required.All before treatments can be deleted.

2.
Funnel plots only work with about 10-20 studies.All funnel plots can be deleted.

3.
Risk of bias based on Cochrane tool is not clear.It should be low risk, unclear risk and high risk.The current description of Yes/No does not provide enough information.

4.
In meta-analysis, "one randomized always analysed" while the numbers in Table 4 are not the numbers in meta-analysis in Review Manager.

5.
The control groups in two studies are different so I am not sure if combining the data from two studies in a meta-analysis is a right when there is such heterogeneity in PICOS 6.Moreover, here enclosed you'll find a copy of the paper.The authors can find some suggestions, as tracked changes and comments to the original manuscript, which could improve the quality of the manuscript.

Are the conclusions drawn adequately supported by the results presented in the review? Yes
If this is a Living Systematic Review, is the 'living' method appropriate and is the search schedule clearly defined and justified?('Living Systematic Review' or a variation of this term should be included in the title.)Yes Competing Interests: No competing interests were disclosed.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have

Figure 2 .
Figure 2. Methodological quality assessment of the 2 studies included in the meta-analysis (0=No/not reported, 1=Yes).

Figure 3 .
Figure 3. Forest plot in defaecation before and after treatment.

Figure 4 .
Figure 4. Funnel plot showing overall standardized mean difference in defaecation before and after treatment.

Figure 5 .
Figure 5. Forest plot in fecal incontinence before and after treatment.

Figure 6 .
Figure 6.Funnel plot showing overall standardized mean difference in fecal incontinence before and after treatment.

Figure 7 .
Figure 7. Forest plot in retentive posturing before and after treatment.

Figure 9 .
Figure 9. Forest plot in severity of pain before and after treatment.

Figure 10 .
Figure 10.Funnel plot showing overall standardized mean difference in severity of pain before and after treatment.

Figure 8 .
Figure 8. Funnel plot showing overall standardized mean difference in retentive posturing before and after treatment.

Figure 11 .
Figure 11.Forest plot in consistency of stool before and after treatment.

Figure 12 .
Figure 12.Funnel plot showing overall standardized mean difference in consistency of stool before and after treatment.

Figure 13 .
Figure 13.Forest plot in acceptance and tolerance after treatment.

Figure 14 .
Figure 14.Funnel plot showing overall standardized mean difference in acceptance and tolerance after treatment.

Reviewer Report 23
May 2019 https://doi.org/10.5256/f1000research.19569.r47237© 2019 Bellini M. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Massimo Bellini Gastrointestinal Unit, Department of Translational Research and New in Medicine and Surgery, University of Pisa, Pisa, Italy I'm really puzzled about the usefulness of a review which takes into consideration only two papers and altogether only 132 patients.Constipation, as the author himself states, is a very frequent condition, so the conclusions are inevitably flawed by this bias.The author should address this issue in the conclusions Hence, I have significant reservations.

Table 3 . Data extraction. Characteristics Seyyed Ali Mozaffarpur Mohammad Reza Esmaeilidooki
Improvement in defecation frequency per week, improvement in the episodes of fecal incontinence per week, improvement in the episodes of retentive posturing per week, improvement in the average of severity of pain of defecation (by VAS), improvement in the average of consistency of stool defecated (by VAS) and patient's drug compliance week, improvement in the average of severity of pain of defecation (by VAS), improvement in the average of consistency of stool defecated (by VAS) and patient's drug compliance).Data extraction was performed by Muhammad Shazad Aslam.Transcripts were analysed, coded and data was extracted using the demo version of qualitative data analysis software Atlas.ti8.0.Table3represent all the data that was extracted.All the meta-data are availableas Dataset 1.
Rome III criteria of functional constipation Rome III criteria of functional constipation Diagnostic test Paraclinics like anorectal manometry, thyroid function tests, anti-tTG, and other laboratory tests thyroid function tests, anti-tTG, and etc.If it remained any doubt, barium study and anorectal manometry would be performed retentive posturing per