The systematic literature review can be characterized as being of medium to high quality. The study demonstrates a rigorous and systematic approach in several aspects, such as the establishment of review methods prior to the conduct of the review, adherence to PRISMA guidelines, a dual-review process for study selection, and the use of the Cochrane risk of bias tool for assessing bias. The authors also conducted an adequate investigation of publication bias and had a plan for exploring heterogeneity. However, the study falls short in certain areas, such as not justifying the exclusion of non-randomized studies, restricting the literature search to articles in English, and not explicitly stating how bias assessment results were used in the synthesis.

Limitations:

Exclusion of Non-randomized Studies: The authors did not provide explicit justification for the exclusion of non-randomized studies. This could limit the scope and applicability of the review findings.

Language Restriction: The literature search was restricted to articles written in English. This could introduce language bias and potentially exclude relevant studies published in other languages.

Lack of Clarity on Bias Assessment Utilization: While the authors used the Cochrane risk of bias tool, they did not explicitly state how the results of the bias assessment were used in the synthesis. This leaves uncertainty regarding the integration of bias assessment in the analysis.

Funding Sources Not Reported: The authors did not report on the sources of funding for the studies included in the review. This is a limitation as funding sources can be associated with the direction and strength of study results, and not reporting them may leave potential conflicts of interest unaddressed.

Details on Data Extraction Process: The authors performed data extraction in duplicate but did not specify the process for resolving disagreements during data extraction, nor did they provide a measure of agreement.

Detailed Information on Included Studies: More specific demographic details about the study participants, as well as details about the setting and location of the studies, would have added more context and depth to the review.

In conclusion, while the study is methodologically sound in several aspects, the aforementioned limitations prevent it from being characterized as high quality without reservations. It is important for readers and users of this systematic review to consider these limitations when interpreting the findings.

Assessment of methodological quality using Amstar 2.0

Domain	Assessment	Justification
Did the research questions and inclusion criteria for the review include the components of PICO?	Yes	The study’s research questions and inclusion criteria incorporate the PICO components adequately.
Did the report of the review contain an explicit statement that the review methods were established prior to the conduct of the review and did the report justify any significant deviations from the protocol?	Yes	The authors explicitly stated that the review methods were established prior to the conduct of the review. They also mention that the review was conducted in accordance with the PRISMA guidelines and registered on the PROSPERO platform. No deviations from the protocol were reported.
Did the review authors explain their selection of the study designs for inclusion in the review?	Yes, but they did not justify the exclusion of non-randomized studies	The authors of the systematic review explained their selection of study designs for inclusion in the review and followed a strategy in selecting study types. However, they did not explicitly justify the exclusion of non-randomized studies.
Did the review authors use a comprehensive literature search strategy?	Partial Yes	The authors thoroughly documented their search strategy, but they only considered articles written in English and did not explicitly mention searching the grey literature.
Did the review authors perform study selection in duplicate?	Yes, but no Kappa score or similar measure was reported	The authors followed a dual-review process for study selection and had a procedure for resolving disagreements. However, they did not report a measure of inter-reviewer reliability.
Did the review authors perform data extraction in duplicate?	Yes, but they did not specify how disagreements were resolved, nor did they provide a Kappa score	The authors performed data extraction in duplicate, but they did not specify the process by which disagreements were resolved, nor did they provide a measure of agreement.
Did the review authors provide a list of excluded studies and justify the exclusions?	Yes	The authors provided a list of excluded studies and justified the exclusions, demonstrating transparency and adherence to systematic review protocols.
Did the review authors describe the included studies in adequate detail?	Yes, but more specific demographic details about the study participants, as well as details about the setting and location of the studies would have been helpful	The authors provided a sufficient level of detail about the included studies and their methodology, but more specific demographic and setting details would have added more context.
Did the review authors use a satisfactory technique for assessing the risk of bias (RoB) in individual studies that were included in the review?	Yes, but it would have been helpful if they had stated how the bias assessment results were used in the synthesis	The authors used the Cochrane risk of bias tool and met most of the criteria for assessing bias. However, they did not explicitly state how the bias assessment results were used in the synthesis.
Did the review authors report on the sources of funding for the studies included in the review?	No	The authors of this systematic review did not report on the sources of funding for the studies included in the review. No mention of funding sources was found in the materials and methods section nor in the results. There is also no discussion regarding the potential impact of funding sources on the outcomes of the studies.
If meta-analysis was performed did the review authors use appropriate methods for statistical combination of results?	Yes	The authors used appropriate methods for the statistical combination of results. They used a random effects model, assessed study heterogeneity, considered the impact of missing data, focused only on randomized controlled trials, assessed reporting bias, and used the Cochrane risk of bias tool for quality assessment.
If meta-analysis was performed, did the review authors assess the potential impact of RoB in individual studies on the results of the meta-analysis or other evidence synthesis?	Yes	The authors assessed the potential impact of risk of bias in individual studies on the results of the meta-analysis. They used the Cochrane risk of bias tool, performed sensitivity analysis when considerable heterogeneity was detected, and conducted an assessment of publication bias.
Did the review authors account for RoB in individual studies when interpreting/discussing the results of the review?	Yes	The authors accounted for the risk of bias in individual studies when interpreting/discussing the results of the review. They used the Cochrane risk of bias tool for quality evaluation, considered the impact of missing data, assessed publication bias, and assessed heterogeneity among studies.
Did the review authors provide a satisfactory explanation for, and discussion of, any heterogeneity observed in the results of the review?	Yes	The authors had a plan to explore heterogeneity, but the actual heterogeneity in their data was null. Therefore, no detailed exploration or explanation for heterogeneity was needed or provided.
If they performed quantitative synthesis did the review authors carry out an adequate investigation of publication bias (small study bias) and discuss its likely impact on the results of the review?	Yes	The authors conducted an adequate investigation of publication bias and discussed its likely impact on the results of the review. They used funnel plots, Begg’s and Egger’s tests, and the ‘trim and fill’ technique to assess and discuss publication bias.
Did the review authors report any potential sources of conflict of interest, including any funding they received for conducting the review?	Yes	The authors reported no potential sources of conflict of interest. The document includes a conflict of interest statement and funding statement.

Detailed assessment of Amstar 2.0 domains

Domain 1: Did the research questions and inclusion criteria for the review include the components of PICO?

Based on the provided text, the study’s research questions and inclusion criteria do seem to incorporate the PICO (Population, Intervention, Control group, and Outcome) components adequately.

Population: The research involves diabetic and non-diabetic patients, both sexes, and all ages and ethnicities, who are either at risk of heart failure, have end-stage kidney disease, or are trying to lose weight.

Intervention: The study clearly states that patients were given SGLT2 inhibitors as part of the treatment. This is the intervention being evaluated.

Control: The control group was those patients receiving a placebo, as stated in the section “Types of interventions administered to patients.”

Outcome: The outcome for this review was the assessment of the prevalence of genital or urinary infections in subjects taking SGLT2 inhibitors compared to those not taking this treatment.

Hence, the study does a good job in articulating all the components of the PICO. Assessment: Yes.

Domain 2: Did the report of the review contain an explicit statement that the review methods were established prior to the conduct of the review and did the report justify any significant deviations from the protocol?

Based on the provided text, the authors explicitly stated that the review methods were established prior to the conduct of the review. They mention that the review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Furthermore, they demonstrate that they had a registered protocol on the PROSPERO platform (registration number: CRD42022375951), indicating a pre-planned methodology and transparent process. This demonstrates independent verification of their written protocol. There is no mention of any significant deviations from the protocol. Thus, based on the provided text, the assessment of this domain is ‘Yes’.

Domain 3: Did the review authors explain their selection of the study designs for inclusion in the review?

Yes, the authors of the systematic review explained their selection of study designs for inclusion in the review and followed a strategy in selecting study types. Specifically, they restricted their search to randomized controlled trials (RCTs), as stated in the “Types of studies” section:

“We considered articles written in English, reporting randomized controlled trials (RCTs) evaluating the side effects of SGLT2 inhibitors, administered to diabetic or non-diabetic patients (i) for prevention of heart failure, (ii) for preserving renal function in end-stage kidney disease or (iii) for weight loss.”

The authors appear to have chosen RCTs because they are the gold standard for clinical research, offering the highest level of evidence. These study types help to minimize biases that could potentially influence the outcomes of the studies, thereby providing the most reliable and valid results. However, the authors did not directly justify the inclusion of only RCTs and the exclusion of non-randomized studies.

The authors also defined clear inclusion criteria concerning the type of participants, interventions, and outcomes they were interested in. Furthermore, they used a systematic approach in their data collection and analysis, assessed study quality and risk of bias, and also performed a statistical analysis.

In conclusion, while the authors did explain their selection of study designs for the systematic review, they did not explicitly justify their restriction to RCTs and exclusion of non-randomized studies. Despite this, given the nature of their research question and the focus on the side effects of a drug, the choice of RCTs is logically understandable.

Domain 4: Did the review authors use a comprehensive literature search strategy?

The authors of the systematic review appear to have employed a comprehensive literature search strategy. They have met the following criteria:

At least two bibliographic databases: The authors searched PubMed and EMBASE.

Years and databases examined: The authors specified they searched records published up to January 31st, 2023, in the aforementioned databases.

Key words and/or MESH terms: The authors used MeSH terms in their search (e.g., [sodium glucose co-transporter-2 OR canagliflozin OR dapagliflozin OR empagliflozin) AND (urinary tract infections OR genital infections OR balanitis OR vulvovaginitis OR candidiasis OR Fournier gangrene)]).

Full search strategy availability: While the authors did not explicitly mention if the full search strategy is available on request, they provided enough detail in the text to reproduce the search.

Supplementing searches: The authors supplemented their search by browsing reference lists of reviews and trial reports, or other sources.

Language: The authors considered only articles written in English.

Grey literature: The authors did not explicitly mention searching the grey literature, which could be a potential limitation.

Given the above points, it seems fair to assess this domain as a “Partial Yes”. The authors have thoroughly documented their search strategy, including the databases and search terms used. However, they have only considered articles written in English and did not explicitly mention searching the grey literature, which may have resulted in some relevant studies being overlooked.

Domain 5: Did the review authors perform study selection in duplicate?

The authors of the systematic review have performed study selection in duplicate. They indicated that the selection of studies was independently conducted by four authors. Two authors screened each database (PubMed and EMBASE) for potentially relevant titles. The authors who screened titles then downloaded relevant ones for full-text reading, with the final inclusion and extraction of relevant information from the studies performed.

When controversies arose in the process of study selection, they were resolved by one independent researcher, which indicates a consensus process being employed when disagreements occurred.

However, the authors do not mention a Kappa score or any quantitative measure for evaluating the agreement between reviewers.

In summary, while the authors followed a dual-review process for study selection and had a procedure for resolving disagreements, they did not provide a Kappa score or similar measure to indicate the level of agreement between reviewers. Thus, the assessment is “Yes” for duplicate study selection but with a note that they could have further strengthened their process by reporting a measure of inter-reviewer reliability.

Domain 6: Did the review authors perform data extraction in duplicate?

The authors of the systematic review appear to have performed data extraction in duplicate. The manuscript states, “Title and abstract screening to exclude documents that did not meet the inclusion criteria was performed independently by four authors (two for each database).” In addition, they mention, “Data extraction was performed by four authors using a standardized form.”

While the authors do not specify the process by which disagreements during the data extraction were resolved, they mention that controversies during the initial screening were resolved by an independent researcher, which suggests that a similar process may have been used during data extraction. However, this is not clearly stated.

Also, the authors did not mention whether a second reviewer checked the agreement on a sample of studies, nor did they provide a Kappa score indicating ‘strong’ agreement (0.80 or greater).

Therefore, the assessment is “Yes” for the authors performing data extraction in duplicate by multiple authors independently. However, the review lacks information on whether a consensus process was used for disagreements during data extraction and if the agreement was checked using a Kappa score.

Domain 7: Did the review authors provide a list of excluded studies and justify the exclusions?

Based on the provided text, the authors of this systematic review did provide a list of excluded studies and justified the exclusions. During the full-text evaluation, 29 articles were excluded for various reasons including: one review article, 15 reports dealing only with type-2 diabetes patients, 2 articles reporting a comparison of SGLT2 inhibitors with other drugs, one letter to the Editor, 5 papers reporting the results presented in other included studies, 2 studies reporting short-term administration of SGLT2 inhibitors, one article dealing with cost-benefit of SGLT2 inhibitors treatment, 3 reports non presenting safety data.

In terms of accessibility, while the authors did not mention whether the list is available on request if it is too long to include in the report, the process of exclusion and the reasons are clearly stated, which indicates a thorough and systematic approach to the review process.

Therefore, the assessment of this domain is “Yes”. The authors provided a list of excluded studies and justified the exclusions. They stated each reason for exclusion, demonstrating transparency and adherence to systematic review protocols. However, the authors could improve the review by making the full list of excluded studies available upon request, if not included in the report due to its length.

Domain 8: Did the review authors describe the included studies in adequate detail?

Based on the provided text, the authors of the systematic review appear to have described the included studies in adequate detail. They have included information on the study design, participants, interventions, outcomes, and have also outlined the search strategy and data extraction methods employed. The authors considered articles reporting randomized controlled trials (RCTs) evaluating the side effects of SGLT2 inhibitors, administered to both diabetic and non-diabetic patients, with the main outcome being the prevalence of genital or urinary infections.

The text outlines the types of patients (participants) involved, mentioning that they were of both sexes and were included irrespective of age or ethnicity. However, more specific details such as the age range of the participants, or their average age, could have been provided to add more context.

The types of interventions administered are described, where patients on treatment with SGLT2 inhibitors or placebo were considered.

The location and setting in which the original studies were conducted are not explicitly mentioned in the provided text. Although this information is often useful for contextual purposes and may impact generalizability, it’s not always critical to the evaluation of a systematic review.

In terms of outcomes, the authors state that the review focused on the assessment of the prevalence of genital or urinary infections in subjects taking SGLT2 inhibitors compared to those not taking this treatment.

The authors provide clear descriptions of their data collection and analysis methods. This includes the search strategy used, selection and extraction of studies, and statistical analysis methods. The authors also explain how they assessed study quality, heterogeneity, and reporting bias.

Additionally, the text includes a “Results” section where the authors discuss the search results and findings of their systematic review and meta-analysis. It also seems that some data has been presented in tables (e.g., Supplementary Table 1) and figures, although these are not included in the provided text.

Overall, the authors seem to provide a sufficient level of detail about the included studies and their methodology. Therefore, the assessment for this domain would be “Yes”. However, it would be helpful if they included more specific demographic details about the study participants, as well as details about the setting and location of the studies.

Domain 9: Did the review authors use a satisfactory technique for assessing the risk of bias (RoB) in individual studies that were included in the review?

Based on the provided text, the authors of this systematic review appear to have used a satisfactory technique for assessing the risk of bias in individual studies that were included in the review. Here’s how:

Use of a formal tool: The authors used the Cochrane risk of bias tool, which is a well-validated instrument for assessing risk of bias in randomized controlled trials.

Risk of bias assessment results for each study: Results of the risk of bias assessment for each study are reported. They categorized the risk of bias for each study as high, low, or of some concerns.

Independent assessors: The authors reported that three authors independently performed the quality assessment, meeting the criteria of at least two independent assessors.

Consensus process for disagreements: The authors mention that disagreements were resolved by discussion, indicating the use of a consensus process when disagreements arose.

Use of bias assessment results in subsequent synthesis: The authors did not use risk of bias to exclude studies, suggesting that the results of the risk of bias assessment were considered in the subsequent synthesis, albeit not to exclude studies but likely to interpret the results. However, it is not explicitly stated how the bias assessment results were used in the synthesis.

Therefore, based on the provided text, the authors seem to have used a satisfactory technique for assessing the risk of bias, meeting most of the criteria, including use of a formal tool, independent assessors, consensus process, and reporting bias assessment results. The only area that could use more explicit clarification is how exactly the bias assessment results were used in the subsequent synthesis, but the provided information does not suggest any unsatisfactory procedures. Hence, the assessment is “Yes”.

Domain 10: Did the review authors report on the sources of funding for the studies included in the review?

Based on the provided text, the authors of this systematic review did not report on the sources of funding for the studies included in the review. There is no mention of funding sources in the materials and methods section nor in the results. Additionally, the authors have not acknowledged that the source of funding for a study can be associated with the direction and strength of the study results. Therefore, the assessment of this domain is ‘No’. Elements missing from this review include the funding sources for the individual studies and any potential conflicts of interest related to those sources. Also, a discussion regarding the potential impact of funding sources on the outcomes of the studies is not included.

Domain 11: If meta-analysis was performed did the review authors use appropriate methods for statistical combination of results?

Based on the provided text, it appears that the authors of the systematic review have used appropriate methods for the statistical combination of results. Here are the main reasons for this conclusion:

Statistical Models: The authors used a random effects model (specifically the Mantel-Haenszel method) for combining the data from the different studies. This is an appropriate choice when there is assumed to be heterogeneity between studies, which is often the case in systematic reviews of clinical studies. The authors also considered the potential impact of heterogeneity by planning a sensitivity analysis if considerable heterogeneity was detected.

Assessment of Heterogeneity: The authors appropriately assessed study heterogeneity using the I^2 statistic, with reference to the established Cochrane criteria for interpretation.

Missing Data: The authors stated that they considered the impact of missing data on the meta-analysis results, which suggests an appropriate handling of missing data.

Combining Different Study Designs: The authors considered randomized controlled trials (RCTs) for their meta-analysis. Since all the studies included were RCTs, no method for combining different study designs was needed.

Assessment of Reporting Bias: The authors assessed publication bias using funnel plots, Begg’s and Egger’s tests, and the ‘trim and fill’ method. These are all standard methods for assessing publication bias.

Quality Assessment: The authors used the Cochrane risk of bias tool to evaluate the quality of the included studies. The risk of bias was not used to exclude studies, but would have informed the interpretation of the results.

The authors seem to have been thorough and used appropriate methods for conducting their meta-analysis. Based on this information, the assessment is ‘Yes’. They used appropriate methods for the statistical combination of results in the meta-analysis.

Domain 12: If meta-analysis was performed, did the review authors assess the potential impact of RoB in individual studies on the results of the meta-analysis or other evidence synthesis?

Assessment: Yes

Justification: The authors of the systematic review have indeed assessed the potential impact of risk of bias in individual studies on the results of the meta-analysis, which aligns with the appropriate guidelines for conducting a systematic review and meta-analysis.

They performed a quality evaluation and risk of bias analysis using the Cochrane risk of bias tool. They considered different types of bias including those associated with the randomization process (D1), deviations from intended interventions (D2), missing outcome data (D3), measurement of outcome (D4), and selection of reported results (D5). The assessment was performed independently by three authors and any disagreements were resolved through discussion. They explicitly stated that the risk of bias was not used to exclude studies, which indicates a comprehensive and inclusive approach to the review.

Further, the authors performed sensitivity analysis when considerable heterogeneity was detected (defined as I2 ≥ 75%), though it’s not explicitly stated whether this included sensitivity analyses excluding studies at high risk of bias. They also conducted an assessment of publication bias, where they used both Begg’s and Egger’s tests, and applied the ‘trim and fill’ missing study imputation approach to address potential bias.

The detailed approach and steps taken by the authors to evaluate the risk of bias and its potential impact on the results of the meta-analysis indicate a rigorous, transparent, and methodologically sound approach. Therefore, it can be inferred that the authors adequately evaluated the potential impact of risk of bias on their meta-analysis results.

Domain 13: Did the review authors account for RoB in individual studies when interpreting/ discussing the results of the review?

Assessment: Yes

The authors of this systematic review have taken multiple measures to account for the risk of bias in individual studies and have considered its potential impact on the findings of the review.

Evidence from the text:

Quality evaluation: The authors used the Cochrane risk of bias tool to assess potential biases, considering multiple aspects like randomization process, deviations from the intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. They further used this information to classify each study as high risk, some concerns, or low risk of bias, which shows a robust and systematic assessment of the risk of bias in individual studies.

Addressing missing data: In cases of missing or insufficient information, the authors considered the impact of missing data on the meta-analysis results. This shows a consideration for the potential bias that missing data might introduce.

Publication bias: The authors assessed publication bias by funnel plot and used Begg’s and Egger’s tests to confirm or exclude the presence of publication bias. They also used the ‘trim and fill’ missing study imputation approach to adjust for potential publication bias, which again indicates a thorough approach to identifying and mitigating bias.

Heterogeneity assessment: The authors assessed heterogeneity among studies using the I2 statistic. This is an important aspect in meta-analyses as substantial heterogeneity can introduce bias in the pooled results.

Although the authors did not explicitly discuss the potential impact of risk of bias on their findings in the ‘Results’ section provided, they clearly included this as a key part of their methods. The systematic and robust methodology used to evaluate and adjust for potential biases implies that these factors would have been considered when interpreting their results.

Domain 14: Did the review authors provide a satisfactory explanation for, and discussion of, any heterogeneity observed in the results of the review?

The authors of the systematic review followed a rigorous methodology, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, assessing the quality of included studies using the Cochrane risk of bias tool, and appropriately handling statistical analyses.

Regarding heterogeneity, the authors used the I^2 statistic to assess the degree of inconsistency between the studies included in their meta-analyses. They also provided threshold interpretations for the I^2 values, which is a good practice. Furthermore, they planned a sensitivity analysis in cases where considerable heterogeneity of pooled analyses including at least 4 studies was detected. This suggests that they had a systematic plan to explore heterogeneity and the robustness of their results.

However, in their meta-analyses, all I^2 values reported were 0%, indicating no heterogeneity. While this is a good sign that the studies were consistent, it also means the authors did not have a chance to further discuss or explore the reasons for heterogeneity as it was not present in their data.

No explicit mention was made of using subgroup analyses or meta-regression techniques to explore potential sources of heterogeneity. However, these tools are often used when substantial heterogeneity is present, which was not the case in this review.

The authors performed an analysis of reporting bias using funnel plots, Begg’s and Egger’s tests, and the ‘trim and fill’ strategy, but these tools are primarily used to assess publication bias and not specifically heterogeneity.

In conclusion, based on the information provided in the text, the authors had a plan to explore heterogeneity, but the actual heterogeneity in their data was null, thus no detailed exploration or explanation for heterogeneity was needed or provided. Therefore, the assessment for this domain is “Yes”, the authors provided a satisfactory plan for, and discussion of, heterogeneity observed in the results of the review.

Domain 15: If they performed quantitative synthesis did the review authors carry out an adequate investigation of publication bias (small study bias) and discuss its likely impact on the results of the review?

The authors of the systematic review conducted an adequate investigation of publication bias, also known as small study bias, and discussed its likely impact on the results of the review. They used appropriate methods to assess this bias, including funnel plots, Begg’s and Egger’s tests, and the ‘trim and fill’ technique, which are standard tools for evaluating potential publication bias in a meta-analysis.

In the Methods section, under “Assessment of reporting bias”, they mention, “Publication bias was assessed by funnel plot in the presence of at least 4 trials in each meta-analysis. If a potential bias was suspected by visual inspection of the plots, the Begg’s and Egger’s tests were used to test funnel plots symmetry and to confirm or exclude the presence of publication bias.” They also mention using the ‘trim and fill’ missing study imputation approach, which helps to estimate the effect of the potential missing studies on the analysis.

In the Results section, they provide specific findings from their investigation of publication bias. They mention that the Begg’s and Egger’s tests were used, and no significant asymmetry was found in most cases, indicating no significant publication bias. In cases where the ‘trim and fill’ strategy was used, they report the adjusted odds ratio after accounting for potentially missing studies. For example, they state, “Although publication bias analysis did not detect a significant asymmetry of the funnel plot (p = 0.31, Egger’s test; p = 0.14, Begg’s test), the ‘trim and fill’ strategy imputed three missing studies on the left side of the funnel plot.”

Thus, the authors not only investigated the presence of publication bias using standard methods, but they also discussed the likely impact of this bias on their results, fulfilling the requirements for an adequate investigation of publication bias. Therefore, the assessment for this domain is “Yes”.

Domain 16: Did the review authors report any potential sources of conflict of interest, including any funding they received for conducting the review?

Based on the provided text, the authors of this systematic review reported no potential sources of conflict of interest. The document includes a conflict of interest statement and funding statement which are necessary elements to evaluate this domain. Therefore, the assessment of this domain is “Yes”.