Risk of diarrhea in patients with type 2 diabetes mellitus treated with sitagliptin: a meta-analysis of 30 randomized clinical trials.

BACKGROUND: Sitagliptin is an important drug used for diabetes treatment and is used as a monotherapy in diabetic patients. However, there are also reported cases of diarrhea with sitagliptin use. Unfortunately, data concerning the relationship of diarrhea with sitagliptin use in various conditions have yet to be identified. Therefore, the overall incidence and risk of diarrhea with sitagliptin use have not been well defined.
METHODS: We conducted searches on Embase, PubMed, and the Cochrane Library databases for relevant randomized controlled trials. Registered relevant trials at the clinical trials registration website were also searched. Statistical analyses were conducted to calculate the overall incidence, odds ratios, and 95% confidence intervals (CI) by using either random-effects or fixed-effect models according to the heterogeneity of the included studies.
RESULTS: A total of 8,891 subjects with diabetes from 30 randomized clinical trials were included in the meta-analysis. The overall incidence of sitagliptin-associated diarrhea was 4.48% (95% CI: 3.59%-5.58%). Compared with the controls, the use of sitagliptin was not associated with a significantly increased risk of diarrhea with an odds ratio of 1.10 (95% CI: 0.78%-1.55%; P=0.58). No evidence of publication bias was observed.
CONCLUSION: Our study has shown that there is no difference in diarrhea risk between sitagliptin and controlled therapies. Moreover, sitagliptin is not a medicine that potentially increases the risk of diabetic diarrhea. More studies are recommended to further investigate this association.

Introduction

Diabetes mellitus (DM) is one of the most prevalent worldwide endocrine system diseases. According to a report by Wild et al1 there were 171 million patients diagnosed with DM worldwide in the year 2000, and this number is predicted to increase to 366 million by the year 2030. Characteristics of DM include hyperglycemia and disturbances in carbohydrate levels and, typically, there is also insulin deficiency and insulin resistance.2 In clinical practice, type 2 diabetes mellitus (T2DM) is more common,3 and the main monitoring indices of T2DM are glycosylated hemoglobin (HbA1c) levels and fasting blood glucose (FBG) levels.4 Thus, the main strategy for T2DM therapy is to control the levels of HbA1c and FBG. At present, oral hypoglycemic agents (OHAs) are the most commonly used treatment for T2DM. Some of these OHAs include sulfonylurea, the biguanide class of drugs, dipeptidyl peptidase-4 (DPP-4) inhibitors, and glucagon-like peptide-1 (GLP-1) agonists.5

Sitagliptin is an important OHA used for T2DM. It was the first DPP-4 inhibitor to be used in clinical practice in 2006.6 DPP-4 is an enzyme located in the endothelium and has a high affinity toward GLP-1 as a substrate.7 Sitagliptin is a highly selective DPP-4 inhibitor that can increase concentrations of intact plasma GLP-1 and lower or normalize blood glucose levels in T2DM patients.8

In monotherapy of T2DM patients, sitagliptin improved HbA1c, FBG, and postprandial glucose with doses of 100 mg and 200 mg once daily.9 FBG was dose-dependently reduced by 17.1 mg/dL (100 mg) and 21.3 mg/dL (200 mg); HbA1c was also reduced by 0.79% and 0.94%, respectively. In another monotherapy study,10 sitagliptin also improved glycemic parameters, including HbA1c and FBG. There are also reports of diarrhea with sitagliptin as monotherapy.9,11 However, data concerning diarrhea with sitagliptin use in different clinical trials have not been evaluated, and the overall incidence and risk of diarrhea with sitagliptin use have not been well defined. As diarrhea could lead to a discontinuation of sitagliptin use and could be fatal in many instances, it is important to fully recognize the risk of diarrhea induced by sitagliptin. Therefore, we conducted this systematic review and meta-analysis to investigate the incidence and risk of diarrhea in T2DM patients treated with sitagliptin.

Methods

Search strategy and study selection

We searched Embase from the dates of 1974 through 2014, PubMed from 1967 through 2014, and the Cochrane Library electronic databases. Specifically, we used the following search terms treated as Medical Subject Headings (MeSH) terms or free text: “diabetes”, “diabetes mellitus”, “type 2 diabetes”, or “noninsulin-dependent diabetes mellitus”; “sitagliptin”; and “randomized controlled trials”, “clinical trials”, “controlled clinical trials”, “clinical trial as topic”, or “randomized controlled trial as topic”. Additionally, we searched the clinical trial registration website (ClinicalTrials.gov) to obtain information on the registered clinical trials. Detailed search strategies are shown in Figures S1–S3. Clinical trials that reported the occurrence of diarrhea with sitagliptin use in adult T2DM patients were eligible for inclusion.

Other inclusion criteria included: 1) the diagnosis of T2DM based on either the American Diabetes Association criteria12 or the World Health Organization criteria,13 2) patients who were assigned to treatment with sitagliptin, and 3) treatment with sitagliptin for a minimum duration of 12 weeks. This duration was chosen based on the fact that a study of this duration could provide relevant information on diarrhea.

Data extraction and quality assessment

Data extraction was conducted by two independent investigators (ZQW and HDS), and studies were retrieved for further consideration if judged pertinent by one or two reviewers. Discrepancies were identified and resolved by consensus or, as needed, by a third investigator (WBH) and confirmed by consensus. When there were multiple reports from the same trial, the most complete and/or most recently reported data were chosen.

For each study, the following information was extracted: first author’s name, year of publication, the Clinical Trials Registry number (No NCT), treatment arm, treatment duration, mean age, number of patients in the treatment and control groups, adverse outcomes (diarrhea), change from baseline of HbA1c and FBG in treatment and control groups, and dosage of sitagliptin. All of the randomized controlled trials (RCTs) included in this review had their quality assessed using the Jadad criteria.14 Scores ranged from 0 to 5, with a high score indicating a high-quality study.

Data analysis

The number of patients with diarrhea in the sitagliptin group and the total number of patients receiving sitagliptin were extracted from the selected trials to calculate the incidence of diarrhea. For each study, the proportion of patients with diarrhea and the 95% confidence interval (CI) were derived. The odds ratio (OR) of diarrhea was calculated only with those assigned to the control group in the same trial. We used the Peto method to calculate the OR and the 95% CI because this method provides the best CI coverage and it was more powerful and relatively less biased when dealing with low event rates.15 Heterogeneity was assessed by using the Q statistic and I2 tests among trials.16 Heterogeneity was considered statistically significant when Pheterogeneity<0.1 or I2>40%. If heterogeneity existed, the data were analyzed using a random-effects model; if heterogeneity did not exist, a fixed-effects model was used. A statistical test with a P-value <0.05 was considered significant. The presence of publication bias was evaluated by using the Begg’s and Egger’s tests.11,17

All data analyses were performed by using Stata software, version 12.0 (Stata Corporation, College Station, TX, USA) and R software, version 3.0.3 (The R foundation for statistical computing, http://www.r-project.org).18

Results

Search results

A total of 1,761 articles and 94 clinical trials were identified initially through our search. After reviewing each study, 1,728 studies were excluded (Figure 1). The remaining 30 studies, with 8,891 subjects, which met our inclusion criteria, were included in our analyses. The basic characteristics of the trials included in the meta-analysis are summarized in Table 1. The quality of the 30 clinical trials was high. Seven of them had Jadad scores of 5,19–25 which described the methods of randomization and blinding appropriately and provided the number of patients who withdrew and dropped from the trials. Thirteen clinical trials had Jadad scores of 4.9,10,26–36 This lower score was due to the fact that the researchers did not describe the methods of randomization or blinding appropriately. Ten studies had Jadad scores of 3, which included seven articles37–43 and three clinical trials from the clinical trial registration website. We performed this meta-analysis in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement44 (Table S1).

Figure 1

Flow chart demonstrating the process of study selection.

Abbreviation: OHA, oral hypoglycemic agent.

Table 1

Baseline characteristics of 30 included trials in the meta-analysis (n=8,891)

First author/sponsor	Year	No NCT	Treatment arm	Mean age (years)	Treatment duration (weeks)	ΔHbA1ca	ΔFPGb	Events of diarrhea	Total events	Jadad score
Charbonnel19	2006	NCT0086515	Sitagliptin 100 mg qd + metformin	54.4	24	−0.67%	0.9 mmol/L	12	464	5
			Placebo + metformin	54.7	24	−0.02%	0.5 mmol/L	6	237
Rosenstock26	2006	NCT00086502	Sitagliptin 100 mg qd + pioglitazone	55.6	24	−0.85%	16.7 mg/dL	3	175	4
			Placebo + pioglitazone	56.9	24	−0.15%	1.0 mg/dL	2	178
Aschner9	2006	NCT00087516	Sitagliptin 100 mg qd	53.4	24	−0.61%	0.7 mmol/L	11	238	4
			Sitagliptin 200 mg qd	54.9	24	−0.76%	0.9 mmol/L	10	250
			Placebo		24	0.18%	0.3 mmol/L	6	253
Raz37	2006	NR	Sitagliptin 100 mg qd	53.6	18	−4.8%	0.7 mmol/L	8	205	3
			Sitagliptin 200 mg qd	55.4	18	−0.36%	0.6 mmol/L	2	206
			Placebo	55.5	18	0.12%	0.12 mmol/L	4	110
Hermansen38	2007	NR	Sitagliptin 100 mg qd + glimepiride ± metformin	55.6	24	−0.45%	−4.4 mg/dL	3	222	3
			Placebo + glimepiride ± metformin	56.5	24	0.28%	15.7 mg/dL	6	219
Goldstein27	2007	NCT00103857	Sitagliptin 100 mg qd	53.3	24	−0.66%	17.5 mg/dL	5	179	4
			Placebo	53.6	24	0.17%	5.8 mg/dL	7	176
Scott45	2008	NCT00541775	Sitagliptin 100 mg qd + metformin	55.2	18	−0.73%	11.7 mg/dL	3	94	4
			Placebo + metformin	55.3	18	−0.22%	6.1 mg/dL	1	91
Chan20	2008	NR	Sitagliptin 50 mg qd	68.9	54	−0.7%	1.0 mmol/L	7	65	5
Raz29	2008	NCT00337610	Sitagliptin 100 mg qd + metformin	53.6	30	−1.0%	1.6 mmol/L	6	96	4
			Placebo + metformin	56.1	30	0	−0.2 mmol/L	5	94
Mohan39	2009	NCT00289848	Sitagliptin 100 mg qd	50.9	18	−0.7%	1.4 mmol/L	0	352	3
			Placebo	50.9	18	0.3%	0.3 mmol/L	0	178
Aschner30	2010	NCT00449930	Sitagliptin 100 mg qd	56.3	24	−0.43%	11.5 mg/dL	19	528	4
Bergenstal31	2010	NCT00637273	Sitagliptin 100 mg qd	52	26	−0.9%	−0.9 mmol/L	16	166	4
Williams–Herman22	2010	NCT00103857	Sitagliptin 100 mg qd	54.1	104	−1.2%	26.8 mg/dL	8	179	5
Barzilai32	2011	NCT00305604	Sitagliptin 100 mg qd	71.6	24	−0.5%	−16 mg/dL	1	102	4
			Placebo	72.1	24	0.2%	11 mg/dL	1	104
Chien40	2011	NR	Sitagliptin 100 mg qd	73.5	24	−1.1%	38.3 mg/dL	0	49	3
Pratley41	2011	NCT00700817	Sitagliptin 100 mg qd	55	52	−0.88%	0.59 mmol/L	14	219	3
Rosenstock42	2012	NCT00642278	Sitagliptin 100 mg qd	51.7	12	−0.74%	12.6 mg/dL	2	65	3
			Placebo	53.3	12	−0.22%	3.6 mg/dL	2	65
Aschner33	2012	NCT00751114	Sitagliptin 100 mg qd	53.3	24	−1.13%	12.4 mmol/L	10	264	4
Russell-Jones10	2012	NCT00676338	Sitagliptin 100 mg qd	52.3	26	−1.15%	−1.1 mmol/L	9	163	4
Bergenstal23	2012	NCT00754988	Sitagliptin 100 mg qd	55.5	156	−0.66%	0.83 mmol/L	10	184	5
Yang34	2012	NCT00813995	Sitagliptin 100 mg qd + metformin	54.1	24	−1.00%	−1.1 mol/L	4	191	4
			Placebo + metformin	55.1	24	−0.10%	0	1	194
Dobs24	2013	NCT00350779	Sitagliptin 100 mg qd + metformin + rosiglitazone	54.4	54	−1.1%	1.7 mmol/L	8	170	5
			Placebo + metformin + rosiglitazone	54.8	54	−0.3	0.6 mmol/L	5	92
Kobayashi43	2014	NR	Sitagliptin 50 mg qd	64.3	24	−0.56%	0.67 mmol/L	2	59	3
Merck Sharp & Dohme Corp62,c	2013	NCT01076088	Sitagliptin 100 mg qd	51.7	24	−0.99%	−21.86 mg/dL	3	120	3
			Placebo	53.6	24	−0.59%	−11.93 mg/dL	4	126
Merck Sharp & Dohme Corp63,c	2014	NCT00411554	Sitagliptin 50 mg qd	60.8	12	−0.7%	19.6 mg/dL	4	163	3
Arjona Ferreira35	2013	NCT00509236	Sitagliptin 25 mg qd	60.5	54	−0.74%	24.7 mg/dL	2	64	4
Merck Sharp & Dohme Corp64,c	2010	NCT00094770	Sitagliptin 100 mg qd	56.8	104	−0.54%	NR	1	588	3
Schernthaner25	2013	NCT01137812	Sitagliptin 100 mg qd	56.6	52	−0.66%	5.85 mg/dL	26	378	5
Lavalle–Gonzalez61	2013	NCT01106677	Sitagliptin 100 mg qd	55.5	52	−0.73%	17.7 mg/dL	23	366	5
Arjona Ferreira36	2013	NCT00509262	Sitagliptin 50 mg/25 mg qd	64.2	54	−0.7%	16.7 mg/dL	11	210	4

Notes:

Value of HbA1c change from baseline.

Value of FBG change from baseline.

Trial date obtained from clinical trial registration website.

Abbreviations: No NCT, number of NCT; n, number of patients; NR, no record; NCT, National Clinical Trial; qd, once a day; HBA, hemoglobin; FBG, fasting blood glucose.

Incidence of diarrhea and subgroup analysis

A total of 5,362 patients who were treated with sitagliptin monotherapy were available for analysis. The incidence of diarrhea ranged from 0% to 10.8%, and the highest incidence occurred in the trials of patients treated with 50 mg of sitagliptin once daily during a 54-week period.20 No events of diarrhea were reported in the two trials.39,40 Based on data from 5,362 patients, the overall incidence of diarrhea was 4.48% (95% CI: 3.59%–5.58%; Figure 2) according to the random-effects model.

Figure 2

Incidence of diarrhea associated with sitagliptin.

Note: aProportion represents diarrhea as a proportion of the different types of reactions to sitagliptin.

Abbreviations: CI, confidence interval; W, weight.

The incidence of diarrhea might be different due to the duration and dosage of sitagliptin treatment; therefore, we performed a subgroup analysis according to the duration and dosage of sitagliptin administration. If the dosage of sitagliptin was ≥100 mg, the incidence of diarrhea for treatment times <30 weeks (3.83%, 95% CI: 3.02%–4.84%; Figure 3) was lower than that for treatment times >30 weeks (5.41%, 95% CI: 3.62%–8.01%; Figure 3). However, if the dosage of sitagliptin was <100 mg, the incidence of diarrhea did not significantly vary between groups with treatment times <30 weeks (4.81%, 95% CI: 1.68%–12.97%; Figure 3) and treatment times >30 weeks (4.83%, 95% CI: 2.83%–8.15%; Figure 3). Through our subgroup analysis, our results suggest that under the conditions of high sitagliptin dosage (≥100 mg), a long treatment time (>30 weeks) might increase the incidence of diarrhea, but low sitagliptin dosages (<100 mg) showed no significant variation between the groups of different treatment times.

Figure 3

Incidence and risk of diarrhea with sitagliptin use based on subgroups.

Note: aProportion represents diarrhea as a proportion of the different types of reactions to sitagliptin.

Abbreviations: CI, confidence interval; W, weight.

ORs of diarrhea

To investigate the specific contribution of sitagliptin to the incidence of diarrhea and exclude the influence of confounding factors such as food, the disease itself, and the history of other therapeutic interventions, we determined the ORs of diarrhea between sitagliptin and placebo groups.9,19,24,26,27,29,32,33,37–39,42,45 The pooled ORs of diarrhea showed that there were no differences in diarrhea risk between treatments with sitagliptin and placebo with an OR of 1.10 (95% CI: 0.78–1.55; P=0.58; Figure 4), according to a fixed-effects model (I2=0%, Pheterogeneity=0.8081).

Figure 4

Odds ratios of diarrhea associated with sitagliptin versus controls.

Abbreviations: OR, odds ratio; CI, confidence interval; W, weight.

Publication bias

We observed no significant evidence of publication bias for the OR of diarrhea events in this analysis by a funnel plot (Figure 5), Egger’s test (P=0.991>0.05, 95% CI: −1.64, 1.62), or Begg’s test (Z=0<1.96, P=1.0>0.05).

Figure 5

Funnel plot of standard error by odds ratio for diarrhea.

Discussion

Diarrhea associated with sitagliptin is an important issue in patients receiving medication. However, due to the high background incidence of diarrhea, it may be very difficult to distinguish between spontaneous and actual drug-related diarrhea events in T2DM patients. Wu et al46 found that, compared with controls, DPP-4 inhibitor monotherapy, which included sitagliptin, was associated with a lower risk of gastrointestinal (GI) adverse events involving diarrhea (relative risk =0.63, 95% CI: [0.55–0.70], P<0.00001). So far, no studies have explored the association between diarrhea and sitagliptin. Therefore, we have conducted this study to determine the overall incidence and risk of diarrhea in T2DM patients treated with sitagliptin.

Sitagliptin as a novel T2DM therapy improves alpha islet function due to the increased concentrations of active GLP-1, which stimulates insulin secretion and inhibits glucagon secretion.47 In one study,48 sitagliptin (100 mg) reduced HbA1c by 0.6% from a baseline of 7.7% in 555 subjects. Another study,49 comprising 743 subjects, reported that sitagliptin reduced HbA1c by 0.8% from a baseline of 7.8%. The improved hyperglycemia during sitagliptin monotherapy was sustained over at least 2 years, as shown in a 52-week study of an open-label extension.21 Sitagliptin at 100 mg reduced HbA1c by 0.7% from a baseline of 7.5%. A placebo-controlled, multicenter study37 showed that sitagliptin at 100 mg or 200 mg once daily reduced HbA1c by 0.6% and 0.5%, respectively, from a baseline of 8.1%. All of these studies have shown that sitagliptin has a positive clinical effect in reducing glycemia as a monotherapy. However, the etiology and pathophysiology of sitagliptin-related diarrhea are largely unknown.

In our study, we strictly defined inclusion criteria to collect the RCTs of T2DM patients treated with sitagliptin. This method of meta-analysis could increase clinical samples and improve productivity based on statistics, and we were able to obtain a more accurate conclusion to choose a suitable therapeutic scheme in clinical practice.

In our study, 8,891 patients from 30 RCTs demonstrated that the pooled incidence of sitagliptin-associated diarrhea was 4.48% (95% CI: 3.59%–5.58%). According to a subgroup analysis, we found that the incidence of diarrhea was related to the duration and dosage of sitagliptin treatment. If the dosage of sitagliptin was ≥100 mg, the incidence of diarrhea with treatment times <30 weeks (3.83%, 95% CI: 3.02%–4.84%) was lower than that with treatment times >30 weeks (5.41%, 95% CI: 3.62%–8.01%).

Of note, we also found no differences in diarrhea risk between sitagliptin and placebo therapies (OR=1.10, 95% CI: 0.78–1.55, P=0.58; Figure 4). Sitagliptin did not increase the risk of diarrhea in patients with T2DM. One possible explanation for this finding is that diarrhea is due to the T2DM itself. As we know, GI disorders are a common complication of T2DM, including abdominal distension, diarrhea, and constipation. One study50 reports that the frequency of diarrhea is 5.3% in T2DM patients. Moreover, a large population survey asking participants about GI-related disorders51 indicated that the OR was 2.06 in diabetic patients compared with controls, and the prevalence of diarrhea was 15.6%. Histological evidence of sympathetic and parasympathetic nerve impairment has been observed in T2DM patients.52 Sympathetic and parasympathetic nerves regulate the movement of the small intestine, and the damaged nerve tracts could impair GI motility. Previous studies have shown similar results with sympathetic nerve selective deletion in the islets of diabetic rats.53 Based on the autonomic nerve changes in T2DM, the risk of diarrhea may easily increase in T2DM patients compared with normal people. Another possible explanation for this finding is the underreporting of diarrhea; in fact, diarrhea as an independent adverse event reported in trials is low. In our research, 30.2% of RCTs were excluded because data on diarrhea were unavailable. Additionally, only 14 RCTs were included in our study to investigate the risk of diarrhea associated with sitagliptin. Thus, the power to investigate the risk is small. Nevertheless, because sitagliptin is increasingly used in routine treatment of T2DM patients and in clinical trials with other agents, endocrinologist and primary care physicians should be aware of the incidence and risk of diarrhea associated with sitagliptin. Furthermore, physicians should monitor and treat this disorder appropriately.

Heterogeneity is a potential confounding factor that can influence the results of all meta-analyses. In the present study, we did not observe a significant heterogeneity in the meta-analysis. However, some limitations did exist. First, the RCTs were performed at various international institutions by different investigators and may have some potential bias in reporting the types of adverse events. In particular, the frequency of diarrhea is underreported in clinical trials independently. Second, the results are from patients at a follow-up time and thus may not be fully reflective of general patients, nor of a more prolonged use. Third, although diarrhea data are prospectively collected in individual studies, there are potentially important differences among the studies, including race, dosage and administration schedules of sitagliptin, duration of the study, and investigators. All of these increase the clinical heterogeneity among included trials. Additionally, our study includes a mixed population of patients treated with sitagliptin-based combination therapy or sitagliptin monotherapy. Therefore, the design of treatment is not the same in all arms, and this might be another source of heterogeneity.

Despite these limitations, the present meta-analysis provides some clinical reference for the treatment of diabetic diarrhea. In the treatment of diabetic diarrhea, it is important to identify the influence of medicine. Metformin, which belongs to the biguanide class of drugs, is the first-line OHA for the treatment of T2DM, and has been shown to reduce the rate of glucose production through reductions in hepatic gluconeogenesis.54,55 However, treatment with metformin is generally associated with a high incidence of adverse GI effects, including diarrhea, nausea, and vomiting.56–58 Diarrhea is the most commonly reported symptom of adverse GI effects, and metformin can lead to malabsorptive diarrhea ranging from 10% to 53% in T2DM patients.59 Furthermore, in 5% of T2DM cases, diarrhea led to the discontinuation of metformin use.60 Our results indicate that sitagliptin did not increase the risk of diabetic diarrhea. The pooled OR of sitagliptin compared with a placebo was 1.10 (95% CI: 0.78–1.55, P=0.58; Figure 4). Sitagliptin is not a potential diarrhea-inducing medicine. If diabetic diarrhea occurred, sitagliptin may be a better choice compared with metformin in T2DM patients.

Conclusion

In conclusion, our study has shown that there is no difference in diarrhea risk between sitagliptin and controlled therapies. Furthermore, sitagliptin is not a medicine that potentially increases the risk of diabetic diarrhea. Further studies are still recommended to investigate this association.

Embase search strategy.

Abbreviation: exp, extensive search.

PubMed search strategy.

Cochrane Library search strategy.

Abbreviations: ab, abstract; kw, keyword; ti, title.

Table S1

PRISMA 2009 checklist

Section/topic	No	Checklist	Reported on page no
Title
Title	1	Identify the report as a systematic review, meta-analysis, or both	1
Abstract
Structured summary	2	Provide a structured summary including, as applicable: background, objectives, data sources, study eligibility criteria, participants, interventions, study appraisal and synthesis methods, results, limitations, conclusions and implications of key findings, systematic review registration number	1
Introduction
Rationale	3	Describe the rationale for the review in the context of what is already known	1
Objectives	4	Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS)	1
Methods
Protocol and registration	5	Indicate if a review protocol exists, if and where it can be accessed (eg, web address), and, if available, provide registration information including registration number	2
Eligibility criteria	6	Specify study characteristics (eg, PICOS, length of follow-up) and report characteristics (eg, years considered, language, publication status) used as criteria for eligibility, giving rationale	2
Information sources	7	Describe all information sources (eg, databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched	2
Search	8	Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated	2
Study selection	9	State the process for selecting studies (ie, screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis)	2
Data collection process	10	Describe method of data extraction from reports (eg, piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators	2
Data items	11	List and define all variables for which data were sought (eg, PICOS, funding sources) and any assumptions and simplifications made	2
Risk of bias in individual studies	12	Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done at the study or outcome level) and how this information is to be used in any data synthesis	2
Summary measures	13	State the principal summary measures (eg, risk ratio, difference in means)	2
Search	8	Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated	2
Risk of bias across studies	15	Specify any assessment of risk of bias that may affect the cumulative evidence (eg, publication bias, selective reporting within studies)	2
Additional analyses	16	Describe methods of additional analyses (eg, sensitivity or subgroup analyses, meta-regression), if done, indicating which were prespecified	2
Results
Study selection	17	Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram	3
Study characteristics	18	For each study, present characteristics for which data were extracted (eg, study size, PICOS, follow-up period) and provide the citations	3
Risk of bias within studies	19	Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12)	3
Results of individual studies	20	For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group and (b) effect estimates and confidence intervals, ideally with a forest plot	3
Synthesis of results	21	Present results of each meta-analysis done, including confidence intervals and measures of consistency	3
Risk of bias across studies	22	Present results of any assessment of risk of bias across studies (see item 15)	4
Additional analysis	23	Give results of additional analyses, if done (eg, sensitivity or subgroup analyses, meta-regression [see item 16])	3
Discussion
Summary of evidence	24	Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (eg, health care providers, users, and policymakers)	4
Limitations	25	Discuss limitations at study and outcome level (eg, risk of bias) and at review level (eg, incomplete retrieval of identified research, reporting bias)	4–5
Conclusions	26	Provide a general interpretation of the results in the context of other evidence, and implications for future research	5
Funding
Funding	27	Describe sources of funding for the systematic review and other support (eg, supply of data); role of funders for the systematic review	5

Reproduced from Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Plos Med. 2009;6(7): e1000097.1

Abbreviations: No, number; PICOS, patient, intervention, comparison, outcome and study; PRISMA, preferred reporting items for systematic reviews and meta-analyses.