key: cord-0969273-alqt8ypa authors: Aj, Beamish; C, Brown; T, Abdelrahman; Harper E, Ryan; Rl, Harries; Rj, Egan title: International surgical guidance for COVID-19: Validation using an international Delphi process - Cross-sectional study date: 2020-06-09 journal: Int J Surg DOI: 10.1016/j.ijsu.2020.06.015 sha: 130df532f494b6848b568a96d936f10a39812e42 doc_id: 969273 cord_uid: alqt8ypa BACKGROUND: International professional bodies have been quick to disseminate initial guidance documents during the COVID-19 pandemic. In the absence of firm evidence, these have been developed by expert committees, limited in participant number. This study aimed to validate international COVID-19 surgical guidance using a rapid Delphi consensus exercise. METHODS: Delphi statements were directly mapped to guidance from surgical professional bodies in the US and Europe (SAGES/EAES), the UK (Joint RCS), and Australasia (RACS), to validate content against international consensus. Agreement from ≥70% participants was determined as consensus agreement. RESULTS: The Delphi exercise was completed by 339 individuals from 41 countries and 52 statements were mapped to the guidance, 47 (90.4%) reaching consensus agreement. Of these, 27 statements were mapped to SAGES/EAES guidance, 21 to the Joint RCS document, and 33 to the RACS document. Within the SAGES/EAES document, 92.9% of items reached consensus agreement (median 89.0%, range 60.5–99.2%), 90.4% within the Joint RCS document (87.6%, 63.4–97.9%), and 90.9% within the RACS document (85.5%, 18.7–98.8%). Statements lacking consensus related to the surgical approach (open vs. laparoscopic), dual consultant operating, separate instrument decontamination, and stoma formation rather than anastomosis. CONCLUSION: Initial surgical COVID-19 guidance from the US, Europe and Australasia was widely supported by an international expert community, although a small number of contentious areas emerged. These findings should be addressed in future guidance iterations, and should stimulate urgent investigation of non-consensus areas. Surgical practice has rapidly changed in response to the novel coronavirus (COVID-19) pandemic. 1 Many hospitals have quickly filled to capacity with non-surgical patients and a substantial, yet poorly understood, risk from COVID-19 to staff and patients has emerged. 2 International professional bodies have been quick to disseminate initial guidance documents during the COVID-19 pandemic. [3] [4] [5] In the absence of firm evidence, 6 and with insufficient time to consult their memberships, guidance has been based predominantly on experiential reports and has been developed by expert committees within these professional bodies; limited in participant number, yet far-reaching in influence. 7 Such leadership, while helpful in aligning early OR practice, is not comprehensive, and cannot hope to integrate the perspectives of the wider medical community. With this approach comes a risk of disproportionate reliance upon a limited perspective, and the representativeness of the guidance produced is unclear. 7 This study aimed to validate the emerging international guidance on surgical practice in COVID-19 against wider international expert consensus. This study was reported in line with STROCSS 8 and is registered online at http://researchregistry.com (UIN: researchregistry5675). 9 Participation in this consensus study was open to all stakeholders in relation to OR practice during the COVID-19 pandemic. Stakeholders were defined as individuals whose work related to the OR (including endoscopy and dentistry), or who had specific knowledge relevant to the novel coronavirus. In a recent rapid systematic review of OR practice in COVID-19, the very limited nature of the literature base was reported, with a clear need for a novel methodological approach during the rapidly emerging pandemic. 6 Four key domains of OR practice emerged; physical resources, personnel factors, patient factors, and procedure-related factors. The earlier systematic review informed the development of the methodology employed in this mapping study of Delphi consensus responses to guidance items. Twitter was chosen as the social media platform for delivery for two reasons. First, because of its ability to allow global distribution and engagement, and second, because of its automatic online cataloguing of responses and ability to flag themes, using the hashtag symbol (e.g. #surgery). This permitted rapid turnaround of a high number of responses in order to expedite project progression. A dedicated Twitter account (@OpCOVID) was created to act as the vehicle for distribution and collection of stakeholder interactions. Phase I -Question collation. Stakeholders were requested, via Twitter, to provide questions relating to the four domains of OR practice during the COVID-19 pandemic. The project was promoted by OpCOVID collaborators via personal Twitter accounts and direct engagement of relevant stakeholders, paying particular attention to relevant, prominent Twitter users. Questions contributed were reviewed by three collaborators (AB, CB, RE), to assess for repetition, themes, and relevance. Phase II -Question answering. Firstly, identified questions were presented to stakeholders via Twitter, with an invitation to report personal and observed practices and experiences, and to signpost relevant evidence and guidance. Meanwhile, collaborators simultaneously interrogated the literature for emerging evidence-based answers and guidance; this evidence was used to stimulate on-going discussion. All responses were extracted, recorded, and the resources identified were downloaded for phase III analysis. During this phase, the same three collaborators (AB, CB, RE) used the deconstructed international guidance items, alongside all other data collected via the social media campaign, to directly inform the development of relevant Delphi statements, in order to ensure effective mapping to the guidance would be possible. Phase III -Consensus exercise. Answers from phase II were used to develop a series of statements within a modified Delphi questionnaire, delivered using Google Forms. Where evidence and/or guidance were identified in phases I and II, these were signposted within the Delphi statement, including links directly to the source(s). A single-round exercise was employed to maximise participation and permit rapid completion. Global Delphi participants were sought via social media emails to stakeholder organisations, and electronic contact with relevant collaborator contacts. Demographic data collected within the Delphi exercise included country of practice, area of expertise, stage of training, and highest academic qualification. A three-point Likert scale was used (one=agree; two=unsure; three=disagree), with an escape option (outside of my expertise) if required. 'Consensus agree' (consensus that the statement was appropriate) was defined as more than 70% of participants allocating an item a score of one AND fewer than 25% of participants allocating a score of three. 'Consensus disagree' (consensus that the statement was not appropriate) was defined as more than 70% of participants allocating an item a score of one AND fewer than 25% of participants allocating a score of three. 'Nonconsensus' was defined as more than 33% of participants allocated an item a score of one AND 33% or more allocating a score of three. All other combinations were considered 'equivocal'. All Delphi statements directly related to a given extracted guidance item were reported. The median (range) agreement was reported and, where any statement was relevant to more than one guidance item, de-duplication ensured it was counted once only in consensus calculations for each guidance document. The Delphi exercise was completed by 339 experts, from 41 countries (Table 1) The only statement not reaching consensus related to decontamination of surgical and endoscopic equipment in cases involving persons suspected or known to have COVID-19. Although not reaching consensus, a majority of 60.4% supported using separate decontamination pathways, the remaining two fifths equally split between being unsure (20.3%), and in disagreement (19.2%). The content of the Joint RCS document is mapped to consensus statements in Table 3 . Consensus was again very high with a median consensus of 87.6% (range 63.4-97.9%). Of 21 statements (de-duplicated from 25), 19 (90.4%) reached consensus agreement. The two statements not reaching consensus both related to operative aspects of surgical practice. Firstly, the preferential selection of an open approach rather than laparoscopy was favoured by a majority of 63.4% participants; 20.7% were unsure, and 15.9% disagreed. The second, a more specific point, was the consideration of stoma formation in preference to gastrointestinal anastomoses to reduce the risk of complications, with 64.0% in support, 20.2% unsure, and 15.8% in disagreement. The content of the RACS document is mapped to consensus statements in Table 4 . Once again consensus was very high with a median consensus of 85.5%, although an outlier led to a wide range (18.7-98.8%). Of 33 statements (de-duplicated from 39), 30 (90.9%) reached consensus agreement. The three statements not reaching consensus related to procedural and personnel aspects. Similarly to the Joint RCS document, the RACS guidance advocated consideration of stoma formation to minimise the risk of resource-consuming complications. As outlined above, a little under two thirds (64.0%) of participants supported this assertion. The second statement failing to reach consensus described the provision of dual attending/ consultant operating. This was mapped to the item recommending allocation of a second consultant or attending in patients with COVID19. A majority of 59.7% participants agreed with dual operating, while 26.2% were unsure, and 14.2% disagreed that it had clinical benefits. The final statement that failed to reach consensus related to participants' understanding of which procedures were aerosol generating. Fewer than one fifth (18.7%) of participants felt that it was clear which procedures generate aerosols, while 20.5% were unsure and 60.5% felt that it was unclear. This study has validated international guidance documents related to surgical practice during the COVID-19 pandemic against broad international expert consensus. The approach engaged 339 worldwide multidisciplinary experts, working in the precise clinical settings of interest, namely the operating rooms in regions affected by COVID-19, making the respondent pool experience-rich. Consensus agreement was evident in 90-93% of statements mapped to the international guidance documents, strongly supporting the initial guidance issued by SAGES/EAES, Joint RCS and RACS. Consensus statements covered four key domains of practice, as identified in a preliminary systematic review: physical resources, personnel factors, patient factors, and procedure-related factors. 6 The 90% of statements achieving consensus covered all four domains. Within the physical resources domain there was clear support for measures to prevent cross-contamination, such as negative pressure operating rooms and separation of COVID and non-COVID areas. Regarding personnel factors there was a clear priority for protecting the staff, participants agreeing upon the use of personal protective equipment (PPE) according to protocols and restriction to essential personnel only in exposed environments. Patient-related factors reaching consensus included discussion of the additional risk of COVID-19 in consent processes, considering all patients to be contagious, and actively screening patients requiring procedures, including the use of thoracic computerised tomography where required. Supported procedural factors were predominated by statements related to restricting workload to essential urgent and emergent work, e.g. prioritising non-operative management where feasible and safe, and reducing potential aerosolisation, e.g. minimising electrosurgical smoke, using filtration devices. The small number of mapped statements failing to reach consensus demonstrates the representativeness of the guidance content in relation to wider expert opinion. Non-consensus statements were predominantly within the procedural domain, illustrating uncertainty regarding the operative approach, the actual risk of aerosolisation of procedures such as laparoscopy, and decisions on stoma formation to reduce post-operative complication risks. In addition, one statement each in the physical and personnel domains, reflected uncertainty over the appropriate way to decontaminate surgical equipment, and the benefit or otherwise of dual consultant operating. Although some evidence has emerged regarding aerosolisation and surface stability of the novel coronavirus pathogen, 10 how these early findings specifically relate to real-world practice remains far from clear. 6, 7 Reliable data influencing surgical practice in the context of COVID-19 remain scarce. Traditional methods of determining best practice underperform in this fast-moving pandemic, 6 and real-time sharing of experience and evidence has taken the lead, while research groups 11 and journals 12 have quickly mobilised to catch up. The guidance documents examined herein are notable in their similarity, addressing themes audible in the corridors of surgical units across the world. The haste with which such guidance has been produced and issued is highly commendable, but beyond this, it is clear from this Delphi exercise that those constructing the documents have very effectively represented the views of this large number of international experts in surgical practice. This study has a number of limitations. Although participants were drawn from all WHO global regions, a majority was based in Europe, so the findings cannot be assumed to be fully representative in the US and Australasia. A single phase Delphi consensus exercise was used, although this compromise permitted the rapid turnaround of this time sensitive project, without reliance upon physical presence from international participants. Two of the three documents examined were issued prior to the consensus process and the documents themselves may have influenced participants' interaction with consensus statements. This Delphi process achieved broad international participation and has validated 90% of the content of early guidance issued by SAGES/EAES, Joint RCS, and RACS on surgical practice in COVID-19. It has demonstrated a clear and consistent ability among multiple professional bodies to produce internationally representative guidance in the early phase of this pandemic. Areas of contention included the surgical approach, dual consultant operating, the separation of decontamination pathways, and consideration of stoma formation rather than anastomosis. These should be carefully considered by guidance issuers, and represent key targets for urgent research. Ethical approval was not required, in accordance with the HRA online tool. This project is registered in the Open Science Framework: https://osf.io/j8zym/ where a protocol can be found. The project is also registered at http://researchregistry.com (UIN: researchregistry5675) The authors report no conflicts of interest. No funding was received for this project. Not commissioned, externally peer-reviewed International surgical guidance for COVID-19: validation using an international Delphi process. Consensus agree: >70% agree, <25% disagree Consensus disagree: >70% disagree, <25% agree Non-consensus: ≥33% agree, ≥33% disagree Equivocal: All other combinations All elective surgical and endoscopic cases should be postponed at the current time. These decisions however should be made locally, based on COVID-19 burden and in the context of medical, logistical and organizational considerations. There are different levels of urgency related to patient needs, and judgment is required to discern between them. However, as the numbers of COVID-19 patients requiring care is expected to escalate over the next few weeks, the surgical care of patients should be limited to those whose needs are imminently life threatening. These may include patients with malignancy that could progress, or with active symptoms that require urgent care. All others should be delayed until after the peak of the pandemic is seen. This minimizes the risk to both, patient and health care team, as well as minimizes utilization of necessary resources, such as beds, ventilators, and personal protective equipment (PPE Only emergency endoscopy and urgent cancer evaluations should be performed during the pandemic 90.6 5.7 3.8 0 Naso-gastric tube placement may be an aerosol generating procedure (AGP). AGPSs are high risk and full PPE is needed. Consider carrying out in a specified location. Where possible, elective surgery should be conducted in hospitals designated as non-COVID or clean sites 81.2 8.1 10.7 0 -339 international experts across 41 countries participated in this Delphi process on surgical practice in COVID-19. -90-93% agreement with guidance from SAGES/EAES (US/Europe), Joint UK Royal Colleges, Royal Australasian College of Surgery. -This study validates the guidance issued by these bodies. -A small number of contentious issues require urgent attention. Please specify the contribution of each author to the paper, e.g. study design, data collections, data analysis, writing. Others, who have contributed in other ways should be listed as contributors. Beamish AJ -Conceived the study and was involved in study design, data collection, data analysis, and writing, including approval of the final manuscript. Brown C, Abdelrahman T, Ryan Harper E, Harries RL, Egan RJ were involved in study design, data collection, data analysis, and writing. All approved the final manuscript and agree to be accountable for its content. All additional collaborators (JA, TE, LH, OJ, SL, WL, OL, KL, DR, RT and AW) were involved in study design, data collection, and critical revision of the manuscript. All approved the final manuscript and agree to be accountable for its content. The Guarantor is the one or more people who accept full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to Name of the registry: Research Registry Unique Identifying number or registration ID Hyperlink to your specific registration The following additional information is required for submission. Please note that failure to respond to these questions/statements will mean your submission will be returned. If you have nothing to declare in any of these categories, then this should be stated. Not required according to the HRA decision tool Please enter the name of the registry, the hyperlink to the registration and the unique identifying number of the study. You can register your research at http://www.researchregistry.com to obtain your UIN if you have not already registered your study. This is mandatory for human studies only.