Efficacy and Safety of STW 5-II for Functional Dyspepsia Treatment: A Patient Data-Based Meta-Analysis

Abstract

Introduction

Functional dyspepsia (FD) is a chronic relapsing gastroduodenal disorder with limited treatment options. Herbal products, like the six-herb combination STW 5-II, can target multiple FD gastrointestinal symptoms. In this meta-analysis, we evaluated the efficacy and safety of STW 5-II for overall FD, and key symptoms, based on Rome IV criteria.

Methods

We systematically screened the literature for randomized controlled clinical studies testing STW 5-II in FD. Meta-analysis was performed using data from individual patients with at least one key FD symptom (fullness, early satiety, or epigastric pain) of at least moderate severity at baseline. ANCOVA-based meta-analyses were performed on improvements in the total symptom sum score, and single symptoms, after 4 and 8 weeks. Safety data were analyzed by calculating odds ratios for all adverse events.

Results

Four randomized controlled trials, including 613 patients, were identified, and two were eligible for efficacy analysis. STW 5-II significantly improved the FD symptom sum score (mean difference of 1.74 after 4 weeks and 2.07 after 8 weeks) and key FD symptoms of fullness (0.28 and 0.29), early satiety (0.25 and 0.26), and epigastric/upper abdominal pain (0.26 and 0.3). Treatment-related or severe adverse events did not differ between STW 5-II and placebo.

Conclusion

The results support that STW 5-II significantly improves FD symptoms after 4 and 8 weeks of treatment with no difference in relation to safety signals compared to placebo. Thus, STW 5-II can be considered an effective and safe treatment option for FD.

Introduction

Functional dyspepsia (FD) is a disorder characterized by chronic or relapsing gastrointestinal (GI) symptoms (GISs) referred to the upper GI tract – including recurring epigastric pain, postprandial fullness, and early satiation – which are not explained by any detected organic cause or biochemical abnormalities [1]. A recent global survey applying Rome IV criteria found an overall global FD prevalence of 7.2% [2], although earlier reports have estimated higher prevalence rates of 10–30% [3–5].

Although the pathophysiology of FD is not completely understood, several potential mechanisms have been proposed – including impaired gastric accommodation, visceral hypersensitivity, Helicobacter pylori infection, altered duodenal microbiome, low-grade mucosal inflammation, and psychosocial factors [1, 6]. FD is not life threatening; however, it affects quality of life and imposes a socioeconomic burden [7–9]. Adequate FD management is important but challenging since most available therapeutic agents only target individual symptoms without enabling overall treatment of FD. Herbal products can offer a more comprehensive approach and thus may play a role in FD management. The Rome IV guidelines list STW 5, an herbal preparation, as a treatment option for FD [1], which is available for decades in many markets. Various alterations of STW 5 have been studied for efficacy over time. Here we focus on the related herbal product STW 5-II containing six herbal extracts: Iberis amara, Carvi fructus, Liquiritiae radix, Matricariae flos, Melissae folium, and Menthae piperitae folium. Pharmacological studies have demonstrated that STW 5-II affects multiple targets, explaining its clinically proven efficacy in relieving FD-related symptoms [10–13]. One investigation showed that STW 5-II has a dual mode of action on gastric motility, i.e., by relaxing the gastric fundus and corpus and toning the antrum [14]. Additional mechanisms of action include the desensitization of hypersensitive afferents [15] and broad anti-inflammatory effects through normalizing histological parameters of inflammation [16] and reducing the levels of several pro-inflammatory chemokines [17–19].

Individual clinical studies have been conducted to evaluate the safety and efficacy of STW 5-II in subjects with FD [10–13], but a meta-analysis has not yet been performed. Here we conducted a meta-analysis to improve the precision and generate new evidence showing the efficacy of STW 5-II for improving overall FD, and individual FD symptoms, according to the Rome IV guidelines, as well as new evidence regarding the safety of STW 5-II in patients with FD.

Materials and Methods

Search Strategy

To ensure the completeness of all clinical data for STW 5-II, a literature search was performed according to PRISMA guidelines, with the expert assistance of a librarian [20]. Literature search including search terms is outlined in a PRISMA flow diagram (Fig. 1).

Fig. 1.

PRISMA flowchart of the study selection process. Studies were identified by searching the electronic databases MEDLINE, Embase, and the Cochrane library, including abstracts from major meetings and conferences. Gray literature was searched with Google and Google Scholar, and the “snowball” method was also utilized to identify all relevant articles. The initial search was not limited to specific languages. The search terms comprised STW 5 II or STW V II or STW5 II or STWV II or STW 5II or STW VII or STW5II or STWVII or IBEROGAST N or IBEROGASTN or iberogast 2 or BAY 98 7410 or BAY98 7410 or BAY 987410 or BAY987410 or IBEROGAST 02 or IBEROGAST02.

Selection of Studies

Abstracts and titles were initially screened independently by a supporting librarian and further by one author (C.F.), and duplicates were eliminated. The full texts of the articles were retrieved, and articles were eliminated if they fulfilled any of the following exclusion criteria: incorrect active pharmaceutical ingredient (including the related herbal medicinal product STW 5), pharmacological study, review, case report, meta-analysis, subgroup analysis, or animal study. Our analysis included clinical studies fulfilling the following inclusion criteria: randomized controlled clinical trials evaluating the efficacy of STW 5-II in adult subjects with FD, studies with clinical endpoints based on reduction of FD symptoms, studies assessing data after at least 28 days of treatment, and studies assessing the safety of STW 5-II.

Risk of bias and heterogeneity (I² statistics) were assessed and additional data on the blinding, funding, study details, patient inclusion criteria, demographic characteristics, and details of the intervention (e.g., duration of treatment and dose) included. Structural comparability between the treatment groups after randomization was verified by comparison of baseline characteristics.

Data Collection Process

All individual patient data from the included studies were available and could be directly transferred into the SAS database. As all data in the publications originated from the original databases, no additional data were retrieved from the publications.

Variables and Endpoints

Efficacy endpoints of primary interest were improvements in global and key dyspeptic symptoms after 28 days and 56 days. The safety objectives were the overall incidence and risk of adverse events (AEs), treatment-emergent adverse events (TEAEs), and frequency of serious adverse events (SAEs).

Statistical Analysis

Full Analysis Set

The full analysis set comprised all patients from the selected studies. Patient data fulfilling the following criteria were selected: at least one moderate key symptom (i.e., score ≥2) at baseline. FD symptom severity was evaluated on a 5-point Likert scale from a validated questionnaire containing 10 GI items (GIS) and outcomes assessed as improvement on days 28 (4 weeks) and 56 (8 weeks). According to the Rome IV criteria [1], the key symptoms of FD are fullness, early satiety, and epigastric or upper abdominal pain. Single symptom analyses included patients with baseline values (>0) available for the respective symptom (Fig. 2).

Fig. 2.

Overview of the included studies for different analyses. The efficacy meta-analysis included only data from placebo-controlled studies.

Data Analysis

The number of patients per treatment group was summarized by study, for all studies pooled, and for all placebo-controlled studies pooled. For efficacy data analysis, improvements in symptom scores from baseline to day 28 (baseline minus day 28) or to day 56 (baseline minus day 56) were submitted for analysis using an SAS^® macro (version 9.4) validated by Cochran Review Manager 5.2. The meta-analysis method was based on adjusted means and standard errors obtained from separate ANCOVAs including the baseline value and treatment as independent variables. The effect size was defined as the difference between adjusted means (least square means) for each endpoint derived from the ANCOVA models. As systematic differences between studies could be ruled out, fixed-effect models were applied to derive common point estimates and associated 95% confidence interval (CIs) with an inverse-variance approach. Cohen’s d [21], as an equivalent to the standardized mean difference (MD), was applied to interpret the importance of the meta-analysis results – with values of <0.2 considered to indicate a negligible treatment effect, 0.5 a medium treatment effect, and 0.8 a large treatment effect. Since all meta-analyses were conducted in exploratory manner, no adjustments of p values were performed [22].

Safety data analyses were performed in SAS (version 9.4; SAS Institute Inc, Cary, NC, USA), including all patients who received at least one dose of the study treatment. Studies without reported events were combined, and one event was assumed in each treatment group. We assessed the overall incidence rates of AEs and TEAEs, and frequency of SAEs, as defined and reported in the original study reports. Meta-analysis was performed using the odds ratio (OR) as the effect size for all AEs and separately for GI AEs for placebo-controlled studies.

Results

Description of Included Studies

Our search strategy identified a total of 102 studies, of which 38 duplicates were removed after the electronic literature search. The remaining 64 studies were screened and 40 excluded for various reasons. Thus, 24 were assessed for eligibility (Fig. 1). A total of four randomized controlled studies (Table 1) fulfilled the eligibility criteria, and the individual patient data were available from the original clinical databases. All four studies were included in the safety meta-analysis [10–13]. Two studies were included in the efficacy meta-analysis [12, 13], after the exclusion of one verum-controlled study [11] and one study utilizing a separate symptom score [10] (Fig. 2). Bias within studies regarding treatment comparisons can be ruled out by evaluating the concealment of randomization, blinding, number of study centers, and funding of the individual studies. Both studies included in the efficacy meta-analysis were randomized with concealed allocation, double blinded, and multi-centric and reported exclusions from analysis data and the reasons for exclusions.

Table 1.

Summary of characteristics of included studies

Study	Overall number of randomized subjects	Study design	Treatment arms	Duration of treatment
Buchert et al. [11] (1994)	243	Double-blind, 3-arm parallel	STW 5-II	4 weeks
			Placebo
			STW 5
Rösch et al. [12] (2002)	180	3-Arm parallel	STW 5-II	4 weeks
			Cisapride
			STW 5
Madisch et al. [13] (2004)	120	Double-blind, cross-over	STW 5-II	8 weeks (+4 weeks)
			Placebo
Vinson and Holtmann [14] (2020)	272	Double-blind, 2-arm parallel	STW 5-II	8 weeks
			Placebo

STW 5-II, herbal product containing six herbal extracts; STW 5, herbal product containing nine herbal extracts.

Analysis Sets of Different Analyses

The full analysis set comprised 613 patients (340 STW 5-II and 273 placebo), including 588 patients fulfilling the Rome IV criteria (323 STW 5-II and 265 placebo). The data set for safety analysis comprised 613 patients with 845 treatment periods (486 STW 5-II and 359 placebo). The STW 5-II and placebo groups had comparable baseline characteristics. The mean age at baseline was 46.3 years (range: 18–78 years) and 62.2% of the patients were female.

The meta-analysis of efficacy data included the two placebo-controlled studies [12, 13], which analyzed the improvements in overall FD severity and FD single symptoms after treatment durations of 28 days and 56 days (Fig. 2). The analysis set for efficacy after 4 weeks included 383 patients (193 STW 5-II and 190 placebo), and after 8 weeks included 326 patients (164 STW 5-II and 162 placebo), with consecutive treatment until days 28 and 56, respectively. Exploratory evaluation of the improvements in overall FD severity and single symptoms after 28 days was based on three studies [11–13] (Fig. 2), and this analysis set comprised 444 patients (254 STW 5-II and 190 placebo).

Efficacy Analysis

Meta-Analysis of Improvement in FD and Single Symptoms

The meta-analysis provided significant results in favor of STW 5-II treatment for the overall improvement in FD after 4 weeks (MD 1.74 [95% CI: 0.90–2.58], p < 0.001) and 8 weeks (MD 2.07 [1.09–3.04], p < 0.001). The results also showed significant improvements in three key FD symptoms after treatment with STW 5-II compared to placebo after 4 weeks and 8 weeks, respectively – including fullness (MD 0.28 [95% CI: 0.10–0.46], p = 0.002 and MD 0.29 [0.09–0.49], p = 0.005), early satiety (MD 0.25 [0.06–0.45], p = 0.011 and MD 0.26 [0.05–0.47], p = 0.014), and epigastric/upper abdominal pain (MD 0.26 [0.10–0.42], p = 0.002 and MD 0.3 [0.10–0.49], p = 0.003) (Fig. 3). Meta-analysis of FD-associated symptoms abdominal cramps, acid eructation/heartburn, loss of appetite, and retrosternal discomfort showed statistical improvement in favor of STW 5-II after 28 and 56 days and in the symptom sickness after 28 days (Table 2).

Fig. 3.

Meta-analysis for improvements in the key symptoms of fullness, early satiety, and epigastric or upper abdominal pain and in the sum score (Madisch [2004] [12] and Vinson [2020] [13]). Patients were only considered for analysis if the symptom was present at baseline (score >0).

Table 2.

Meta-analysis results for FD symptoms with regard to improvements with STW 5-II versus placebo

GIS item	After 28 days			After 56 days
	MD, 95% CI	p value	heterogeneity test p value (I2), %	MD, 95% CI	p value	heterogeneity test p value (I2), %
Epigastric pain	0.26 (0.10–0.42)	0.002*	0.001 (90)	0.3 (0.10–0.49)	0.003*	<0.001 (93)
Fullness	0.28 (0.10–0.46)	0.002*	0.006 (87)	0.29 (0.09–0.49)	0.005*	0.003 (89)
Early satiety	0.25 (0.06–0.45)	0.011*	0.121 (58)	0.26 (0.05–0.47)	0.014*	0.061 (72)
Nausea	0.11 (−0.10 to 0.31)	0.300	0.057 (72)	0.13 (−0.06 to 0.33)	0.185	<0.001 (93)
Vomiting	−0.23 (−0.53 to 0.06)	0.120	0.069 (70)	−0.24 (−0.56 to 0.09)	0.153	0.163 (49)
Abdominal cramps	0.22 (0.04–0.39)	0.013*	0.015 (83)	0.33 (0.15–0.51)	<0.001*	0.002 (90)
Acid eructation/heartburn	0.30 (0.10–0.50)	0.003*	0.027 (80)	0.34 (0.13–0.55)	0.002*	0.002 (89)
Sickness	0.20 (0.00–0.40)	0.047*	0.265 (20)	0.16 (−0.09–0.41)	0.202	0.119 (59)
Loss of appetite	0.23 (0.01–0.45)	0.038*	0.207 (37)	0.30 (0.05–0.54)	0.016*	0.198 (40)
Retrosternal discomfort	0.33 (0.14–0.52)	<0.001*	0.336 (0)	0.28 (0.05–0.52)	0.019*	0.065 (71)
Sum score	1.74 (0.90–2.58)	<0.001*	0.003 (88)	2.07 (1.09–3.04)	<0.001*	<0.001 (94)

CI, confidence interval; GIS, gastrointestinal symptom; I², i-square test; MD, mean difference.

Analyzing the MDs between STW 5-II and placebo regarding the change in symptom scores for fullness, early satiety, and epigastric or upper abdominal pain revealed Cohen’s d values between 0.30–0.35. Additionally, the Cohen’s d values were approximately 0.4 for change in the sum score at both days 28 and 56.

Supportive Analysis of Changes in Overall FD and Key Symptoms from Baseline to Day 28

Exploratory ANCOVAs based on data from the three placebo- and verum-controlled studies [11–13] showed similar results. After 28 days, the mean improvement and standard deviation (SD) in the GISs score was 7.4 (SD 5.19) with STW 5-II and 4.8 (SD 4.72) with placebo (MD 2.44 [95% CI: 1.57–3.3], p < 0.0001).

For fullness, after 28 days, the mean improvement in the single symptom score was 1.3 (SD 1.01) with STW 5-II and 0.9 (SD 0.91) with placebo (MD 0.38 [95% CI: 0.20–0.56], p < 0.0001). For early satiety, after 28 days, the mean improvement in the single symptom score was 1.1 (SD 0.98) with STW 5-II and 0.7 (SD 0.98) with placebo (MD 0.36 [95% CI: 0.17–0.55], p = 0.0002). For epigastric or upper abdominal pain, after 28 days, the mean improvement in the single symptom score was 1.3 (SD 0.94) with STW 5-II and 0.9 (SD 0.89) with placebo (MD 0.39 [95% CI: 0.23–0.55], p < 0.0001). For the FD symptoms abdominal cramps, acid eructation/heartburn, loss of appetite, retrosternal discomfort, nausea, and sickness, ANCOVA revealed significant improvement in favor of STW 5-II after 28 days of treatment (Table 3). Statistical analysis did not reveal any effects of gender, or any significant differences between age-groups, concerning improvements in the GISs.

Table 3.

ANCOVA results for FD symptoms with regard to improvements with STW 5-II versus placebo after 28 days

GIS item	MD (95% CI)	p value
Epigastric pain	0.39 (0.32–0.55)	<0.0001*
Fullness	0.38 (0.20–0.56)	<0.0001*
Early satiety	0.36 (0.17–0.55)	0.0002*
Nausea	0.28 (0.08–0.47)	0.0071*
Vomiting	0.06 (−0.25 to 0.37)	0.6950
Abdominal cramps	0.32 (0.14–0.50)	0.0005*
Acid eructation/heartburn	0.45 (0.25–0.65)	<0.0001*
Sickness	0.27 (0.08–0.46)	0.0052*
Loss of appetite	0.33 (0.13–0.54)	0.0018*
Retrosternal discomfort	0.52 (0.32–0.71)	<0.0001*
Sum score	2.44 (1.57–3.3)	<0.0001*

CI, confidence interval; GIS, gastrointestinal symptom.

Safety Analysis

The safety evaluation was based on all 845 treatment periods. The incidence rates of TEAEs are based on all available data. The meta-analysis of the placebo-controlled studies revealed that the ORs were consistently below unity (<1), descriptively indicating a lower risk of any AE during STW 5-II treatment compared to placebo. For the different groups of AEs, the raw incidence rates and ORs were as follows: for all TEAEs, 10.1% with STW 5-II versus 13.9% with placebo (OR 0.68 [95% CI: 0.42–1.12]); for GI TEAEs, 2.3% with STW 5-II versus 3.6% with placebo (OR 0.78 [95% CI: 0.32–1.92]); for drug-related TEAEs, 2.5% with STW 5-II versus 1.9% with placebo (OR 0.69 [95% CI: 0.22–2.17]); and for drug-related GI TEAEs, 0.8% with both STW 5-II and placebo (OR 0.66 [95% CI: 0.11–3.81]). Heterogeneity between studies was not indicated (all tests revealed p values >0.48). Overall, three SAEs were reported – including one (0.3%) with STW 5-II treatment (cardiovascular problems) and two (0.7%) with placebo (psychiatric decompensation and anemia). None of these SAEs were reported as being drug related.

Discussion

This is the first meta-analysis to assess the efficacy and safety of STW 5-II in patients with FD meeting various iterations of the Rome criteria. It provides further evidence on the relief of key FD symptoms in patients that meet the criteria of the various iterations of the Rome criteria for FD. The present analysis was based on individual patient data derived from the clinical databases of prior studies that tested STW 5-II in FD. Completeness of the clinical data was ensured by performing a literature review, which revealed no additional published clinical trials investigating STW 5-II in FD. A recently published systematic review and meta-analysis on herbal medications did not specifically focus on the efficacy of STW 5-II in FD [23]. The main inclusion criteria were FD symptoms consistent with the Rome IV criteria, comprising at least one of the key symptoms of fullness, early satiety, and upper abdominal pain, rated as moderate. These criteria were met by 95% of the STW 5-II patients and 97.1% of the placebo patients from the original study population.

The meta-analysis results demonstrated significant improvements in overall FD symptom burden (p < 0.001) and in the three key symptoms of FD (p < 0.02), after 28 and 56 days. The standardized MD of 0.3–0.35 indicates a treatment effect of moderate strength. The symptoms of nausea, vomiting, and sickness did not show a significant difference between STW 5-II and placebo. These symptoms are accompanying symptoms of FD, but do not reflect the most characteristic or bothersome symptoms of FD [1, 24], as displayed by the very low baseline values for these symptoms in our cohort. The supportive exploratory analysis including one further verum-controlled study [11] substantiates the evidence of the superiority of STW 5-II over placebo with regard to the improvement of nausea and sickness (p < 0.01) after 28 days of treatment. This finding was mainly due to an increased treatment effect of STW 5-II in the verum-controlled study.

Overall, while no full diagnosis according to Rome IV criteria is retrospectively possible, the data selection for the meta-analysis can be considered as a reasonably accurate reflection of FD patients as per definition of Rome IV in clinical practice. Data extraction and statistical analysis were performed using established techniques for meta-analysis; however, one primary limitation of our study is the small number of included studies restricting an overall generalizability of the results and a significant heterogeneity for several symptoms (p < 0.05) detected. A variety of factors influence placebo-response rates [25] and subsequent response to therapy. The heterogeneity could be due to the inherent limitations of the studies included in this meta-analysis including study design and setting (e.g., primary vs. secondary or tertiary care) [26]. The heterogeneity does not limit the validity of the combined effects since all study-specific results revealed the superiority of STW 5-II over placebo. No prospective study protocol was generated for these meta-analyses. However, the statistical analysis plan included all relevant aspects of Cochran-compliant analysis, including specification of the objective, definition of endpoints, mapping of severity scores, and description of statistical methods. While this meta-analysis used a formalized search of relevant databases, all data that were used for the meta-analysis were derived from the clinical databases. Thus, data were not simply extracted from publications that would have been subject to review by a second person.

In general, herbal products are popular and one of the treatment options for functional GI disorders. A similar herbal combination product, STW 5, has been widely used for the treatment of FD, irritable bowel syndrome, and gastritis, with proven efficacy and safety in clinical studies. However, STW 5 was previously the subject of safety discussions [27]. A very rare idiosyncratic liver effect is assumed in post-marketing data. General previous discussions on celandine (a component of STW 5) as in high doses a potentially hepatotoxic ingredient are reflected by the safety evaluation of the German health authority for celandine containing medicinal products due to liver toxicity in 2005 [28]. However, evaluation of all of the toxicological, non-clinical, clinical, and post-marketing data regarding STW 5 did not identify any mechanism of action overall or in relation to celandine. In the full set of ICH-compliant toxicological studies, a high safety margin was demonstrated for STW 5 with doses 600–1,200 times higher than the recommended daily dose. No hepatotoxic effects have been observed in any of the clinical trials.

Nevertheless, this discussion prompted scientific interest and evaluation of STW 5-II, an herbal product similar to STW 5 but lacking three components: Silybum marianum (milk thistle), Angelica archangelica (angelica), and Chelidonium majus (celandine). The present meta-analysis demonstrated the clinical efficacy of STW 5-II in FD and also focused on the safety of this product. For STW 5-II, the meta-analysis of the safety parameters revealed a favorable safety profile comparable to placebo in all four clinical trials performed for the indication of FD. Compared to placebo, treatment with STW 5-II was not associated with any increased risk of AEs. This is further supported by a clinical trial investigating STW 5-II in patients with irritable bowel syndrome, which also did not reveal AEs or SAEs in the STW 5-II group [12]. Moreover, the results of toxicological tests with the high safety margins have also been applied for STW 5-II. Although only limited experience from post-marketing data is available for STW 5-II, no safety issues are expected. With regard to the six components of STW 5-II and their respective concentrations within the formulation (Iberis amara, Carum carvi, Matricaria chamomilla, Melissa officinalis, Mentha piperita, and Glycyrrhiza glabra), no safety issues have been described in either the scientific literature or monographs of the European Medicines Agency.

In summary, for the safe and efficacious treatment of FD, phytotherapy offers a holistic treatment approach by targeting numerous underlying modes of action. The results of the present meta-analysis demonstrate that STW 5-II improves both overall FD and key symptoms of FD after 4 and 8 weeks of treatment and can be considered another safe treatment option for this difficult-to-treat disorder.

Acknowledgment

The authors thank Turacoz Healthcare Solutions (turacoz.com) for writing assistance on an early draft of the manuscript.

Statement of Ethics

Ethical approval was not required because this study retrieved and analyzed data from already anonymized data sets of studies, which were subject of ethics approvals.

Conflict of Interest Statement

V.A.: speakers and/or consulting fees from Arena, Bayer, Falk, Ferring, Hexal, KyowaKirin, 4M-Medical, Medice, Sanofi, Schwabe, and Shionogi. A.S.: no conflicts of interest to declare. C.F. and S.R.: employees of Steigerwald Arzneimittelwerk GmbH. M.W.: funding from Steigerwald Arzneimittelwerk GmbH to conduct the analysis. G.H.: advisory boards for Australian Biotherapeutics, Glutagen, and Bayer; research support from Bayer, Abbott, Pfizer, Janssen, Takeda, and Allergan; boards of West Moreton Hospital and Health Service, Queensland, UQ Healthcare, Brisbane, and the Gastro-Liga, Germany; a patent for the Brisbane aseptic biopsy device; editor of the Gastro-Liga Newsletter; and a member of the Research Committee of the Royal Australasian College of Physicians.

Funding Sources

Steigerwald Arzneimittelwerk GmbH funded this meta-analysis and engaged in conception and design of work and drafting and revising the manuscript.

Author Contributions

V.A.: interpretation of data and drafting, conceptualization, and critical revision of the article. A.S.: critically revising the draft for important intellectual content. C.F.: interpretation of data and drafting and critical revision of the article. S.R.: conception and design of the work, acquisition of funds, interpretation of data, and critically revising the draft. M.W.: conception and design of the work, analysis of the data, and critical revision of the draft. G.H.: interpretation of data and critically revising the draft for important intellectual content.

Funding Statement

Steigerwald Arzneimittelwerk GmbH funded this meta-analysis and engaged in conception and design of work and drafting and revising the manuscript.

Data Availability Statement

Availability of the data underlying this publication will be determined according to Bayer’s commitment to the EFPIA/PhRMA “principles for responsible clinical trial data sharing.” This pertains to scope, timepoint, and process of data access. As such, Bayer commits to sharing upon request from qualified scientific and medical researchers patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the USA and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 1, 2014. Interested researchers can use www.vivli.org to request access to anonymized patient-level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the member section of the portal. Data access will be granted to anonymized patient-level data, protocols, and clinical study reports after approval by an independent scientific review panel. Bayer is not involved in the decisions made by the independent review panel. Bayer will take all necessary measures to ensure that patient privacy is safeguarded.