key: cord-1043003-vqbax34i
authors: Vuichard-Gysin, Danielle; Sommerstein, Rami; Kronenberg, Andreas; Buetti, Niccolò; Eder, Marcus; Piezzi, Vanja; Gardiol, Céline; Schlegel, Matthias; Harbarth, Stephan; Widmer, Andreas
title: High adherence to national IPC guidelines as key to sustainable VRE control in Swiss hospitals: a cross-sectional survey
date: 2022-01-28
journal: Antimicrob Resist Infect Control
DOI: 10.1186/s13756-022-01051-9
sha: a29cf979eddcccc47338d4d7fae4125b1ec72d0b
doc_id: 1043003
cord_uid: vqbax34i

BACKGROUND: Vancomycin resistant enterococci (VRE) are on the rise in many European hospitals. In 2018, Switzerland experienced its largest nosocomial VRE outbreak. The national center for infection prevention (Swissnoso) elaborated recommendations for controlling this outbreak and published guidelines to prevent epidemic and endemic VRE spread. The primary goal of this study was to evaluate adherence to this new guideline and its potential impact on the VRE epidemiology in Swiss acute care hospitals. METHODS: In March 2020, Swissnoso distributed a survey among all Swiss acute care hospitals. The level of adherence as well as changes of infection prevention and control (IPC) strategies in the years 2018 and 2019 after publication of the national guidelines were asked along with an inventory on VRE surveillance and outbreaks. RESULTS: Data of 97/146 (66%) participants were available, representing 81.6% of all acute care beds operated in Switzerland in 2019. The vast majority—72/81 (88%) responding hospitals—have entirely or largely adopted our new national guideline. 38/51 (74.5%) hospitals which experienced VRE cases were significantly more likely to have changed their IPC strategies than those 19/38 (50%) hospitals without VRE cases p = 0.017). The new IPC guidelines included (1) introduction of targeted admission screening in 89.5%, (2) screening of close contacts of VRE cases in 56%, and (3) contact precaution for suspected VRE cases 58% of these hospitals. 52 (54%) hospitals reported 569 new VRE cases in 2018 including 14 bacteremia, and 472 new cases in 2019 with 10 bacteremia. The ten largest outbreaks encountered between 2018 and 2019 included 671 VRE cases, of which most (93.4%) consisted of colonization events, 29 (4.3%) infections and 15 (2.2%) bacteremia. CONCLUSION: Wide adoption of this VRE control guideline seemed to have a positive effect on VRE containment in Swiss acute care hospitals over two years, even if its long-term impact on the VRE epidemiology remains to be evaluated. Broad dissemination and strict implementation of a uniform national guideline may therefore serve as model for other countries to fight VRE epidemics on a national level. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13756-022-01051-9.

Vancomycin resistant enterococci (VRE) are increasing in many European countries. In their last annual epidemiological report, the European Center for Disease Control and Prevention has issued a statement of Open Access *Correspondence: danielle.vuichard-gysin@swissnoso.ch † Stephan Harbarth and Andreas Widmer contributed equally to this work 1 Swiss Center for Infection Prevention, Swissnoso, Bern, Switzerland Full list of author information is available at the end of the article Page 2 of 10 Vuichard-Gysin et al. Antimicrobial Resistance & Infection Control (2022) 11:19 concern [1]. In 2018, the Federal Office of Public Health (FOPH) commissioned Swissnoso, the Swiss Center for Infection Prevention, to investigate a large countrywide nosocomial outbreak of VRE affecting several hospitals with spill-over into many other cantons [2] . This investigation revealed gaps in national surveillance and communication between hospitals and public health authorities. This prompted the Swiss Federal office of public health to making reporting of VRE clusters mandatory [3, 4] .

There is no evidence from any European country, that current strains of VRE and clonal complexes that are endemic in healthcare settings, have started to circulate in the community [5] . This contrasts with ESBLproducing E. coli, which are introduced into healthcare settings through widespread transmission in the community [5] . Lack of adherence to infection control measures, including extensive environmental contamination and high antibiotic pressure, seem to be major drivers of in-hospital VRE acquisition and transmission [6, 7] . VRE transmission mainly results in colonization of the patient, but rarely causes invasive infections such as bacteremia [8, 9] . Therefore, transmission in the absence of systematic screening policies will remain undetected allowing VRE to spread within and across healthcare facilities [10] [11] [12] . That strict adherence to contact precautions (CP) can prevent local transmission of multidrug-resistant organisms, including VRE, was recently demonstrated by a local Swiss healthcare institution [13] . However, compared to other European countries such as France [14] , Germany [15] , or the Netherlands [16] , Switzerland had no uniform national guidelines for prevention and control of VRE transmission in hospitals. Therefore, the national center for infection prevention (Swissnoso) issued new guidelines in September 2018, and updated them in December 2019 [17, 18] , based on scientific evidence and existing international and national publications [19, 20] . The goal was not only to eliminate the clusters responsible of the outbreak, but to largely control VRE at the national level. In addition to the mandatory reporting, Swissnoso made use of the systematic surveillance implemented by the national center of antibiotic resistance (ANRESIS) and started publishing quarterly data on new VRE cases reported to this platform (www. anres is. ch) [21] .

The main objective of this study was to evaluate the adherence with the new national VRE control guideline and its potential impact on the VRE epidemiology in Swiss acute care hospitals. To improve our national VRE surveillance strategy in the future, our secondary goal was to externally validate the VRE data collected by ANRESIS.

In March 2020, a 34-item questionnaire was sent to 146 infection control professionals at 204 acute-care institutions providing inpatient care. Long-term care facilities, nursing homes, and psychiatric institutions were excluded. Non-responding institutions were reminded three times by e-mail.

A questionnaire from a survey conducted in 2018 was updated to meet the current needs [12] . The questionnaire was pre-tested by several infection control nurses and physicians for comprehensibility and time needed. The survey was translated in the three official languages and eventually shared through the online platform Sur-veyMonkey ® . Participants were asked to provide answers for their institutions as well as for others they provided IPC services. If respondents indicated that they were answering for more than one center, they were required to indicate whether those answers were the same or different with respect to the IPC measures. If they were different, we asked them to complete the survey separately for each center.

We collected information on local VRE control strategies including self-reported level of adoption of the guidelines, recent changes in IPC measures, presence of local standards for VRE screening and preventive CP, and whether there were any barriers to implement specific recommendations. In addition, we asked hospitals to provide their total number of new VRE cases detected by year and the proportion of VRE positive blood cultures detected. VRE was defined as Enterococcus faecium with phenotypical amoxicillin-and vancomycin-resistance on routine susceptibility testing. VRE cases were considered "new" if the first detection occurred at the corresponding institution. For hospitals reporting outbreaks, we requested a description of the largest outbreak including detailed information such as the number of isolates detected overall and in blood cultures, resistance phenotypes, and the mode of detection (culture and/or PCR). In addition, we asked them to provide a rough estimate of the clonal relatedness of isolates, e.g., whether < 50%, 50-75% or > 75% were considered clonally related. An outbreak was defined as a situation with an unexpected accumulation of ≥ 3 cases with a positive laboratory test result for VRE from either a clinical sample or screening specimen and with an epidemiological link (temporal, local). Molecular genetic detection was not necessarily required to meet the definition. Survey results were analyzed respondent-based. The survey allowed for multiple attempts to enter data by the same hospital. In case of conflicting answers between different versions, the latest data entry was considered. In addition, the first author contacted the participants by e-mail and asked to verify their entries. For the preparation of this manuscript, the authors followed the revised standards for quality improvement reporting excellence (SQUIRE 2.0) [22] .

Data were exported from the online platform to an Excel ® spread sheet, checked for accuracy, cleaned, and imported for descriptive analyzes into SPSS ® [25]. Categorical data were compared by Chi-square test. Results were either stratified by hospital size or by cantons, as deemed suitable.

The Swiss Centre for Antibiotic Resistance (ANRESIS) is a national surveillance system. Participating clinical microbiology laboratories report their antimicrobial susceptibility test results anonymously. The laboratories are well distributed across the different territories in Switzerland and represent isolates from tertiary-care hospitals, as well as cantonal and private laboratories. All antimicrobial resistance data are derived from routinely performed analyses and include isolates from sterile as well as non-sterile sites. Resistance reports are publicly available with data aggregated by predefined regions [21] .

One of the co-authors (AK) performed an extraction of all VRE cases reported between January 1st, 2018, and December 31st, 2019, and classified them by canton. The surveillance period covered 81% and 89% of annual patient-days (PDs) for 2018 and 2019, respectively. Using the annual VRE numbers from the survey and the ANRESIS database for 2018 and 2019, which were both aggregated at the cantonal level, we created Bland-Altman diagrams for each, bacteremia and nonbacteremia, separately [23] and estimated the correlation coefficient. We defined a priori that agreement between the two datasets was high, if 90% or more of the data points (indicating the difference between the numbers reported in the survey and those retrieved from ANRESIS) clustered around the mean of the differences within two standard deviations of the mean. Agreement was defined as moderate, if at least 75% (but < 90%) of the data points lie within the 95% limits of agreement. Each data point represents the data of a canton.

In addition, we evaluated agreement between the two surveillance tools by visual inspection of the bar charts showing the absolute numbers of new VRE cases by years and cantons separately for blood cultures (bacteremia) and other clinical samples (non-bacteremia) according to Smith et al. [24] .

We addressed a total of 146 infection control professionals or hospital epidemiologists who are responsible for 204 acute care sites. Of these, 97 (66%) participated in the survey, 81 representing a single institution and 16 being responsible for several sites, translating into a total of 116 acute care sites or institutions across 24 cantons. The participants indicated to serve a total of 22,106 beds, which is 81.6% of all acute care beds operated in Switzerland in 2019 according to the Swiss Federal Office of Public Health [25] . The participants represented 63 (65%) small (< 200 beds), 25 (26%) medium (200-500 beds) and 9 (9%) large (> 500 beds) hospitals (Fig. 1) .

Overall, 88% (72/81) of the responding IPC professionals stated that they had fully or largely adopted the Swissnoso guideline in their institutions. Stratified by hospital size, these were 86% (44/51), 90% (19/21), and 89% (8/9) of all responding small, middle and large hospitals, respectively (Fig. 2) . Of 90 respondents, 57 (63%) indicated an intensification of their infection prevention and control (IPC) strategies since beginning of 2018. Among the different hospital sizes, the proportion of medium-sized hospitals that enhanced their VRE control measures was highest at 78.3%, compared with 57.9% (small hospitals) and 66.7% (large hospitals). However, this difference was not statistically significant (Table 1 ). In contrast, there was a significant association between VRE epidemiology and intensification of IPC management: Of 51 hospitals which noted VRE cases, 38 (74.5%) enhanced their VRE control measures within the last two years compared to 19 of 38 hospitals (50.0%) with no VRE cases (p = 0.017) ( Table 1) . Changes in IPC strategies included introduction of admission screening (51/57, 89.5%), contact precaution of confirmed VRE cases (18/57, 31.6%), preventive CP for suspected VRE cases (33/57, 58%), screening of close contacts of VRE cases (32/57, 56.1%), and intensification of environmental decontamination (19/57, 33.3%). Among the different hospital sizes, the proportion of hospitals that introduced admission screening was especially high in small-and middle-sized hospitals (90.9% and 94.4%, respectively), while introduction of CP for confirmed VRE cases was a novum for 39.4% of small hospitals and introduction of preventive CP for VRE suspects was new for 83.3% of large hospitals (Fig. 3) . As far as the knowledge of the guideline was concerned, 77 (88.5%) were aware of the 2019 update and 7 (8%) were aware of at least the original Swissnoso recommendation issued in 2018. A total of 87 IPC professionals responded to this question. The inventory on measures for VRE prevention and control already in place in hospitals according to the Swissnoso recommendations revealed a high compliance with the recommended measures across all hospital sizes. With respect to specific recommendations for targeted admission screening, most small, medium, and large hospitals indicated to screen all transfers from hospitals in other cantons with an ongoing VRE epidemic as well as direct transfers from hospitals abroad. The proportion of hospitals with a recommendation to screen patients with a previous hospital stay abroad or for transfers from highrisk wards such as haemato-oncology was lower, whereas the proportion of hospitals with a written standard for screening patients with a history of VRE carriage markedly varied according to hospital size with the smallest proportion among medium-sized hospitals (Additional file 1: Fig. S1a ). The survey further revealed that > 90% of hospitals had a written standard for preventive CP. The weights put on specific recommendations, however, varied in a similar way as for admission screening. While most hospitals indicated to have recommendations for preventive CP for VRE contacts, transfers from other cantons with a VRE epidemic and for direct transfers from abroad, the proportion of respondents was lower regarding recommending preventive CP for patients with a previous stay abroad, transfer from another ICU with unknown epidemiologic situation or a high-risk ward such as haemato-oncology (Additional file 1: Fig. S1b) (Fig. 4) . The numbers reported by hospitals also varied considerably between the federal cantons (Additional file 1: Fig.  S2 ). Bern was the canton most affected by the large VRE outbreak as has been previously described [2] . From 2018 to 2019, there was an overall increase of VRE detection by means of admission screening with a strong proportional increase from 32 to 89% among small hospitals (Additional file 1: Fig. S3 ). The Proportions of bacteremia remained low in both years with 14 (2.5%) in 2018 and 10 (2.1%) in 2019.

A total of 14 facilities reported at least one outbreak (3 of each small and medium-sized hospitals and 8 large hospitals), 8 in 2018 and 6 in 2019, resulting in a total of 16 outbreaks over the two years. Of all outbreaks reported, the ten largest outbreaks were described in more detail and characterized by a median duration of 14 weeks (interquartile range, IQR 6.5-37.0); three took place each on intensive care units, on surgical wards, and internal medicine wards, two occurred on neonatology, one on geriatrics, and one outbreak affected several wards not otherwise specified. 

Visual inspection of the bar charts showing the absolute numbers of new VRE cases by year and canton for bacteremia (Fig. 5a ) and non-bacteremia (Fig. 5c) showed good concordance between the two reporting systems (survey and ANRESIS). Comparison of the number of new VRE isolates as reported by survey participants and recorded in the ANRESIS database stratified by canton showed a strong correlation for bacteremia (correlation coefficient of 0.88) and non-bacteremia (correlation coefficient 0.98) isolates. Agreement between the two reporting systems was high with 32/34 (94%) measured differences in bacteremia and 33/34 (97%) measured differences in nonbacteremia lying between the 95% limits of agreement, which is above the pre-defined value of 90% (Fig. 5b, d) .

The outlier in Fig. 5d with a difference of 84 belongs to the canton Bern that was most affected by the largest outbreak Switzerland has ever experienced.

This survey reflects the first comprehensive inventory on the strategies to prevent and control the spread of VRE within and across hospitals in Switzerland and an update of the current VRE epidemiology including the detailed description of the ten largest outbreaks. To our surprise, 88% (72/81) of the responding IPC professionals indicated to have fully or largely adopted the Swissnoso guideline in their institutions. There was no significant difference between the size of hospitals regarding guideline adoption. However, we observed that hospitals who had already encountered VRE cases were significantly more likely to have recently enhanced their VRE control strategy. Our results show that a high proportion of Swiss acute care hospitals have stringent measures in place to control the spread of VRE, thus exhibiting a high level of compliance with the national recommendations published by Swissnoso. Furthermore, we did not encounter significant barriers to the implementation of these guidelines. A national survey across German hospitals revealed that most of them had guidelines in place to prevent transmission of and/or infections with methicillin-resistant Staphylococcus aureus (MRSA) (99%), multidrug-resistant Gram-negative pathogens (96%) and Clostridioides difficile (96%) [26] . Results on adherence to recommendations for prevention and control of VRE transmission were not reported. Except for specific outbreak containment measures that have been previously addressed by several co-authors of this manuscript [12] , we did not identify any other publications assessing adherence to VRE control guidelines at national level, at least not in European countries. This work could therefore serve as a good example of a rational approach to assessing IPC compliance beyond Switzerland, applied also to other multidrug-resistant organisms.

We have previously suggested that varying implementation of measures to control the spread of multi-drug resistant organisms in hospitals and different applications of the concept of CP [27] are potential reasons why evidence from studies on the effectiveness of these measures are largely lacking [28] . Compared to countries that resigned from strict VRE containment a while ago [29] , our data suggest, that most cantons in Switzerland are successful in controlling local transmission and stopping the inter-cantonal spread of VRE between hospitals due to measures such as admission screening and preventive CP.

Compared to previously published data from Switzerland [12] , the number of new VRE cases detected per year, however, has more than doubled. In addition to expanded admission screening in small and medium hospitals, the main reason for this sharp increase in newly diagnosed VRE cases was the large-scale outbreak in the Canton of Bern and surrounding regions, which continued into 2019 and spread to the north-east of Switzerland. Despite this increase in VRE cases, the proportion of invasive isolates remains at a low level. This is in contrast with surveillance data from neighboring countries [30] , which show a substantial increase in VRE bloodstream infections, causing a completely different VRE burden. This lower rate of bacteremia could be an indication that our strict measures have an overall containment effect and that the dissemination of VRE in Swiss hospitals has not yet reached an endemic level where intensive control strategies become ineffective. After all, we have seen a sharp decline in VRE cases in large hospitals which were responsible for the main VRE burden. Previous work has indicated that certain clones of E. faecium have a particular propensity for causing nosocomial outbreaks [31, 32] . Thus, even if our results indicate that ANRESIS is a valuable passive surveillance tool, systematic collection of sequencing data may be the crucial next step for the national prospective VRE surveillance in Switzerland and may further contribute to the knowledge about the effectiveness of specific interventions.

Multi-locus sequence typing, pulsed-field gel electrophoresis, and amplified fragment length polymorphism have been proven to be valuable and cheap methods to establish clonality between isolates in different outbreak investigations [33] . However, for bacteria with a more complex genetical diversity such as VRE, whole genome sequencing (WGS) seems to be superior to the conventional typing methods due to its higher discriminative power [34] [35] [36] . WGS may also enhance understanding of intra-and inter-regional spread provided that data are collected at the national level which is the case for example in Denmark [37] .

One of the strengths of our work is the comprehensiveness of the survey addressed to a broad IPC community. The results of which enabled to accurately describe the specific IPC measures established in Swiss acute hospitals and the current VRE epidemiology.

Limitations include the moderate response rate, which was potentially related to the ongoing COVID-19 pandemic with a shift of resources toward outbreak management. In addition, the granularity of reporting VRE cases by canton can be criticized, since canton-level data may be difficult to be interpreted due to heterogeneity in terms of number and size of hospitals, case mix and total population. Nevertheless, in a federal state with competent public health authorities in each canton, these figures can help to better understand the local epidemiology Only cantons with at least one VRE case have been considered. b Bland-Altman Plots for the level of agreement between the two surveillances in reporting cases with VRE bacteremia. "diff" indicates the absolute difference in reported cases between the two systems, "lower" indicates lower limit of agreement, "upper" indicates upper limit of agreement, while "mean" indicates the mean of differences, also called bias. c Total number of new non-bacteremia VRE cases reported by hospitals and recorded by ANRESIS per canton and year (2018-2019). The blue bars correspond to the numbers reported in the national survey, the orange bars correspond to the numbers collected by ANRESIS. Only cantons with at least one VRE case have been considered. d Bland-Altman Plots for the level of agreement between the two surveillances in reporting non-bacteremia VRE cases. "diff" indicates the absolute difference in reported cases between the two systems, "lower" indicates lower limit of agreement, "upper" indicates upper limit of agreement, while "mean" indicates the mean of differences, also called bias and guide appropriate IPC measures. Finally, the period of two years for the comparison of observations is too short to adequately validate the ANRESIS VRE reporting and determine whether the bias and agreement observed are consistent.

The VRE landscape in Swiss acute hospitals has changed significantly in recent years, with an upward trend similar to many other countries in Europe. Nevertheless, the high adherence to the national VRE guidelines gives hope that this negative trend can be halted for the time being. ANRESIS has proven to be a valuable monitoring tool to inform hospitals about the epidemiological situation in other cantons and may therefore serve as additional help to guide local IPC measures.

European Centre for Disease Prevention and Control. Antimicrobial resistance in the EU/EEA (EARS-Net). Annual epidemiological report

Outbreak of vancomycin-resistant Enterococcus faecium clone ST796

Successful collaboration between Swissnoso and the Federeal Office of Public Health during a national vancomycin resistant enterococcus faecium outbreak

Revision der Verordnung des EDI über die Meldung von Beobachtungen übertragbarer krankheiten des Menschen: Wichtigste Anpassung per 1

Vancomycin-resistant enterococci: Why are they here, and where do they come from?

Factors associated with Vancomycin-resistant Enterococcus acquisition during a large outbreak

Effects of antibiotics on nosocomial epidemiology of vancomycin-resistant enterococci

Risk of vancomycin-resistant Enterococcus (VRE) bloodstream infection among patients colonized with VRE

Recommendations for the successful control of a large outbreak of vancomycin-resistant Enterococcus faecium in a non-endemic hospital setting

Factors associated with acquisition of vancomycin-resistant enterococci (VRE) in roommate contacts of patients colonized or infected with VRE in a tertiary care hospital

Comparison of three PCR primer sets for identification of vanB gene carriage in feces and correlation with carriage of vancomycin-resistant enterococci: interference by vanB-containing anaerobic bacilli

Emergence of vancomycin-resistant enterococci in Switzerland: a nation-wide survey

Does high adherence to contact precautions lead to low in-hospital transmission of multidrug-resistant microorganisms in the endemic setting?

Twenty years of antimicrobial resistance control programme in a regional multi hospital institution, with focus on emerging bacteria (VRE and CPE)

Control of the spread of vancomycin-resistant enterococci in hospitals: epidemiology and clinical relevance

Dutch guideline for preventing nosocomial transmission of highly resistant microorganisms (HRMO)

Eindämmung der Verbreitung von Vancomycin-resistenten Enterokokken (VRE) in der Schweiz. Aktualisierung der nationalen Empfehlungen, Version 2.0 (Dezember

Maitriser la propagation des entérocoques résistants à la vancomycine (VRE) en Suisse: mise à recommnedations nationales. Version 2.0 (Décembre

Control of an outbreak of vancomycin-resistant enterococci in several hospitals of western Switzerland

Prévention de la transmission croisée des Bactéries Hautement Résistantes aux antibiotiques émergentes (BHRe)

SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process

Measuring agreement in method comparison studies

Bar charts enhance Bland-Altman plots when value ranges are limited

Infection control and safety culture in German hospitals

Variability in contact precautions to control the nosocomial spread of multi-drug resistant organisms in the endemic setting: a multinational cross-sectional survey

Infection control and prevention measures to reduce the spread of vancomycin-resistant enterococci in hospitalized patients: a systematic review and meta-analysis

Thirty years of VRE in Germany-"expect the unexpected": the view from the National Reference Centre for Staphylococci and Enterococci

Continuous increase of vancomycin resistance in enterococci causing nosocomial infections in Germany-10 years of surveillance

Genomic insights to control the emergence of vancomycin-resistant enterococci

Vancomycin-resistant Enterococcus faecium sequence type 796-rapid international dissemination of a new epidemic clone

How to: molecular investigation of a hospital outbreak

Amplified fragment length polymorphism and whole genome sequencing: a comparison of methods in the investigation of a nosocomial outbreak with vancomycin resistant enterococci

Whole-genome sequencing revealed independent emergence of vancomycin-resistant Enterococcus faecium causing sequential outbreaks over 3 years in a tertiary care hospital

Hospital outbreak of linezolid-resistant and vancomycin-resistant ST80 Enterococcus faecium harbouring an optrA-encoding conjugative plasmid investigated by whole-genome sequencing

Multiple hospital outbreaks of vanA Enterococcus faecium in Denmark

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations

The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13756-022-01051-9.Additional file 1. This file contains the supplementary figures S1a. and b., S2 and S3.Authors' contributions DV, RS, SH, and AW conceived the study. DV, RS, NB, and VP designed and elaborated the survey. DV collected, cleaned, and analyzed the data and wrote the first draft of the manuscript. AK provided the data from ANRESIS and double checked the validation results. RS, AK, NB, ME, CG, MS, SH, and AW substantially contributed to the interpretation of the data and provided critical inputs to the manuscript. All authors read and approved the final manuscript.

This work was partially funded by the Swiss Federal Office of Public Health. Dr. Buetti is currently receiving a grant from the Swiss national Science Foundation (Grant No. P4P4PM_194449).

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

This study was performed as quality improvement project that does not require approval by the ethics committee as it did not involve individual patient data or patient groups.

In this study, no individually identifiable data are included. All data are presented as pooled data. As such, consent for publication was not required.

The authors declare that they have no competing interests.