Background

Live broadcast of surgical procedures (LBSP) has gained popularity in the past few decades and is commonly used in educational events such as conferences and live digital learning events. It is regarded as an opportunity for experts to demonstrate novel techniques and emerging technology especially in complex procedures for the benefit of a targeted audience [1].

This has been particularly applied to the field of Minimally Invasive Surgery (MIS) such as robotic, laparoscopic and or endoscopic techniques as the operative field can be live streamed to a large audience. It can also be an opportunity to see how an experienced surgeon would deal with intraoperative complications, including the decision making process of how they respond to unexpected findings/events. This has been commonly used within the context of educational conferences and events where narration provided allows the audience to engage and interact with the experts through the moderator(s).

There have been, however, some patient safety and ethical concerns during these live surgery broadcast [2]. Surgeons will often travel to perform LBSP outside their normal institutions, which may result in ‘jet lag’ and added fatigue that can alter their performance. Importantly, these surgeries are often complex techniques and or emerging technology which can be technically demanding, adding more stress on the surgical team. Additionally, surgeons often work with potentially unfamiliar teams and equipment which may add further anxiety to the performing surgeon, who may already feel under more pressure to perform in front of a large audience.

Furthermore, the extra personnel required to conduct the live transmission may contribute to distraction to the operating team and may affect the level of infection control in the theatre environment. Finally, expert surgeons do not routinely follow-up those patients and so will be reliant on the host team to care for them postoperatively. Given the likelihood complexity of those cases, host teams may be not familiar with the associated risks and complications.

An evidence synthesis was performed in 2014 looking at the educational value and patient safety of LBSP [3]. They found a comparable rate of complications between procedures performed as LBSP compared to non-LBSP. However, the success rate in some procedures was lower in LBSP group, as demonstrated in endoscopic completion rate. Since the last review, there has been a rapid uptake of MIS across all specialities with more reports on the application LBSP in surgical training. This, however, has been significantly impacted by the unprecedented COVID-19 pandemic with the cancellation of national and international meetings [4]. Alternative ways of remote teaching such as webinars have been widely adopted during the pandemic, but it is unclear if LBSP should be an integral part of this virtual teaching. Therefore, it was necessarily to update the evidence synthesis and re- appraise the literature.

The focus of this review is patient safety issues related to LBSP in Minimally Invasive Surgery (MIS), but a more global view on the matter is also addressed such as ethics and confidentiality.

Methods

Search strategy

A comprehensive review of the literature was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines [5]. The PubMed, Embase and Medline electronic databases were used and a search of the data from 1 March 2005 till 20 February 2021 was performed. The start date was chosen to reflect some of the earliest substantial reports on LBSP in the literature (excluding case reports). Search terms used combinations of “broadcast”, “live surgery”, “safety”, “ethic” and “live surgical broadcast” separated by the Boolean operator “AND”, were used. The search was performed independently by two separate authors. A further search was performed on 10 October 2020 with search terms “COVID-19”, “live surgery”, “live surgical broadcast”, in order to capture any reported literature during the pandemic to date. A subsequent manual search of the related study references was conducted in order to capture reports that were missed in the initial search, and all identified suitable studies were included. Abstracts and conference entries were excluded at initial screening due to the inability of capturing complete sets of data.

Study selection

Studies were eligible for inclusion if they were written in the English language and available in full text. More specifically the following inclusion criteria were used: randomised controlled studies, controlled trials or cohort studies reporting the outcomes of LBSP among all surgical specialties; studies that reported broadcast in conferences as well as other educational events and the guidelines on LBSP across the different surgical bodies and institutions were also included.

There was no restriction on the type of MIS that was conducted and a wide range of procedures were included.

Studies were excluded if the total number of cases used was less than ten. All duplicate studies or studies using the same data set were excluded. All studies using paediatric patients were not included. Only studies reporting outcomes of procedures and complications were included, studies just reporting on delivery of LBSP or ethics were excluded. Finally, reports on social media transmission that were not broadcast live were also excluded.

Outcome measures

The number of patients, demographics (age and sex), type of surgery and reported outcomes and complications were recorded. This included success of procedure (there was heterogeneity in the reporting of this outcome dependant on the surgery performed), return to theatre and readmissions. Complications reported were compared to those patients who were not involved in LBSP.

Additionally, data on ethics, were also captured including dedicated consent forms for LBSP, maintain confidentiality through the live surgery broadcast and reporting on outcomes of those procedures.

Data extraction and analysis

Data extraction was performed independently by two separate authors (MA and MC). After the initial screening of titles and abstracts, articles which fulfilled the eligibility criteria were identified and full text accessed for further analysis.

The methodological quality of the studies was assessed using the Risk of Bias in Non-randomised Studies of Interventions tool (ROBINS-1).

The results are described using a narrative analysis, and primarily grouped by the outcome assessed. Meta-analysis could not be performed on the data due to the heterogeneity of the studies as most of the studies were not statistically comparable.

Results

Demographics

The search strategy identified 1230 abstracts. After removal of duplicates and exclusions were made, 27 final manuscripts were selected for data extraction, including thirteen original papers and fourteen guidelines/position statements. The PRISMA diagram is summarised in Fig. 1.

Fig. 1
figure 1

PRISMA Diagram

Full size image

Eight papers reported on events performed at a single centre while five were multicentre studies in the analysed 13 full text papers [6,7,8,9,10,11,12,13,14,15,16,17,18]. Eight papers reported on events live streamed from a European country, with the remainder being from the USA and Asia. Only two of the papers reported on the number of attendees at their event [8, 18] (Table 1).

Table 1 Summary of included studies [6,7,8,9,10,11,12,13,14,15,16,17,18]
Full size table

Eight papers were retrospective unadjusted cohort studies [8,9,10,11, 15,16,17,18], four retrospective case matched [6, 12,13,14] and one prospective case matched [7]. They addressed a wide range of specialities including urology, endoscopy, upper gastrointestinal surgery and cardiac interventions.

Patient safety outcomes

Of the 13 papers included, nine reported no difference in the complications nor the rate of success of the procedures performed (Table 1).

Four papers reported inferior outcomes in live broadcast. Ramirez-Backhaus et al. looked into the outcomes of LBSP in laparoscopic radical prostatectomy [7]. 23 patients were broadcasted live compared to 46 matched controls. They found no difference in the rate of perioperative or postoperative complications. However, they reported a significant difference in the rate of positive surgical margins in the LBSP group (43.5% vs 17.4%).

Ruiz de Gordejuela et al. analysed data obtained from bariatric courses, where LBSP was performed over a ten-year period [8]. Procedures performed included Laparoscopic Roux-en-Y gastric bypass, sleeve gastrectomy as well as endoscopic procedures. These were compared to patients undergoing similar procedures, not broadcast live. However, they were not matched controls. They found higher complication rates in LBSP and higher rate of re-operation needed. In the LBSP group, 13% of patients experienced complications as follows: Clavien type 1 (4 cases), Clavien type II (3 cases) and Clavien type III (6 cases) with bleeding being the most common complication (72.4%). This is compared to 6.7% in the group not performed as LBSP.

The two other studies reported inferior outcomes related to completion of the procedures during Endoscopic Retrograde Cholangio pancreatography (ERCP). Liao et al. investigated ERCPs broadcast live and compared them to matched controls across 36 conferences [13]. There was no statistically significant difference in the rate of complications between patients in LBSP and control (10.3% vs. 8.6%). However, there was a lower success rate in the live broadcast group (94.1% vs. 97.5%). Ridtitid et al. similarly searched the difference in outcomes and success rate in 82 risk-stratified ERCPs performed during live broadcast compared to matched controls [14]. The complexity of the ERCP was divided as per the grading proposed by Cotton et al. [19]. Patients who had level 1 and 2 procedures were classified as the standard group and patients who had level 3 and 4 procedures were classified as the complex group. They found that in technically simpler procedures there was no significant difference in the complications or success rate between LBSP and matched controls. However, the success rate was significantly lower in those procedures performed live compared to controls (73% vs. 90%; OR, 0.3; 95% CI 0.14–0.69) in more complex ERCPs. The overall success rate in the LBSP group was lower than controls (81% vs. 91%; OR, 0.4; 95% CI 0.21–0.85), but the complication rates were overall similar (6% vs. 3%; OR, 2.2; 95% CI 0.68–7.19).

Legemate et al. reported on the use of LBSP in ureterorenoscopic (URS) and percutaneous nephrolithotomy (PNL) urinary stone procedures and compared them to patients not performed as LBSP (unmatched). No statistically significant difference in the complication rate was found when compared to routine surgical practice. However, they found that the anaesthetic time was significantly longer in the LBSP group undergoing ureterorenoscopic procedures. They also found the retreatment rate was higher in this group. Subsequent multiple logistic regression analyses revealed that there is no association between LBSP and inferior outcomes (after adjusting for confounding variables) and so the authors concluded that LBSP does not compromise patient safety [17].

Ethics, confidentially and consenting

Dedicated consent forms were reported to be used in only 9 studies [6,7,8,9,10,11, 14,15,16]. One study reported on a pooled sample of data across several events and there heterogeneity in reporting dedicated consent [18]. In those that included a dedicated consent process, there was a lack of clarity of the contents of the consent form for LBSP.

Four of the 13 studies did not expand how they maintained ethics and confidentiality [7, 12, 15, 17] during the live surgery broadcast and 5 papers only mentioned that either their study was approved by the local ethics committee or they adhere to their associated guidelines, without further clarification.

Analysis of LBSP guidelines

14 guidelines were identified spanning across different surgical associations, representing four cardiothoracic, one urology, five endoscopy and one from ophthalmology. The remaining three were published by Royal Colleges of Surgeons of England and Australia and The Royal Australian and New Zealand College of Ophthalmologists [1, 20,21,22,23,24,25,26,27,28,29,30,31,32].

All guidelines and position statements addressed the issue of patient safety (Table 2). The majority are in agreement that a moderator must be present in order to facilitate the interaction between the operator and the audience. This is in order to allow the operator to fully focus on the procedure and avoid distraction. There is unanimous agreement that if is felt that continued broadcasting will have an adverse effect on the patient, it should be immediately terminated. There was, however, no proposed framework on a reporting mechanism of patient outcomes during and following LBSP except by the European Association of Urology [1].

Table 2 Summary of the guidelines [1, 20,21,22,23,24,25,26,27,28,29,30,31,32]
Full size table

All guidelines encouraged familiarity with the location and equipment. Therefore, some recommend that where possible the surgeon should perform the procedure from their ‘home’ institution. Where this is not possible, they should familiarise themselves with the environment beforehand and should be working with a team highly familiar with the workplace. They should submit preferences for equipment beforehand in order to avoid unfamiliarity during the procedure. Personnel in theatre should be kept to a minimum and non-clinical staff must not interfere with the procedure.

There is also unanimous agreement among all the published guidelines for the need to have a dedicated consent process for LBSP. The consent should include discussion about the risks of LBSP to the patient such as increased surgeon distraction and the possible breaches of confidentiality. Also the consent should declare that there are potentially limited benefits to the patient to participate in LBSP. In a proportion of guidelines, there is a necessity for the operating surgeon or physician to meet and be involved in the consenting process with the patient in person before the procedure.

All guidelines stated that all attempts need to be made to protect patient confidentiality during and after the procedure. This involves the protection of patient identifiable information appearing on monitors and screens inadvertently.

The patient needs to at all times be reassured that they may refuse or withdraw their consent at any point without their care being affected. If their procedure is delayed due to withdrawal of consent, all attempts must be made to re-arrange their procedure in a timely manner. There must be no coercion of the patient.

Quality assessment

Using the ROBINS-I tool, eight of the studies had an overall moderate risk of bias and five studies had a serious risk of bias (Fig. 2).

Fig. 2
figure 2

Risk of bias (ROBINS-1)

Full size image

Educational value

There was a paucity of original objective data reporting the educational value among the papers.

In their synthesis, Brunckhorst et al. addressed the educational value of LBSP. They note the lack of objective data on the subject. The majority of data reported is through subjective surveys showing participants rating LBSP highly as an educational tool. Only one study reported by Brunckhorst et al. addressed the validity of LBSP objectively. McIntyre et al. compared students in operating theatres against students watching LBSP. In the LBSP group, students asked four times as many questions compared to the group in the operating theatre and had fewer questions unanswered [3].

From the included studies in this synthesis, none measured objectively the educational value.

Discussion

Live surgical broadcast is now widely used among the different surgical specialities due to the perceived educational value of these events. Patient safety concerns however have been raised over the outcomes of patients during LBSP, owing to the different circumstances in which their surgery is being performed. Given the global impact of the pandemic on educational events, the role of live surgery broadcast within the context of virtual training curricula became more unclear. We felt it was appropriate to re-visit this topic, update the literature on LBSP and summarise the outcomes and the wider issue of ethics and patient safety in relation to MIS.

Educational value

Although the literature in this review highlighted the potential educational value of LBSP, it was difficult to draw a conclusion on the magnitude of benefits as this often was not objectively measured and was subjectively reported via surveys. Attempts have been made to do this via questionnaires after events by Skouras et al. [33], which was not included in the analysis as they did not report on clinical outcomes. They conducted surveys to examine the educational value of plastics and ENT live and or pre-recorded surgical events. They found that participants consistently reported that live events were of more educational value than pre-recorded videos. Participants felt that they acquired practical tips which they can apply to their own practice to a greater extent than in pre-recorded videos. Also, the interactive nature of the live surgery meant that they were more alert and engaged throughout. In contrast, Legemate et al. explored the views of urologists and participants in an educational event where LBSP was used, as well as pre-recorded unedited videos. Participants felt overall the education value was similar and a substantial percentage of the surgeons performing the LBSP reported that they felt that live surgery did not provide optimal circumstances for patient safety. Therefore, the authors advocated the use of more pre-recorded unedited videos [34].

Patient safety

Patient safety was the main focus of this review and it was addressed by the success of the procedure and or the complication rate. In this review, we found almost one-third of the papers reported inferior outcomes with LBSP such as less favourable oncological outcomes [7], failure to complete the procedure endoscopically (partial success in 35 cases [8.6%] and complete failure in 24 cases [5.9%] as reported by Liao et al.; partial success in 6 cases [12%] and failure in 8 cases [15%], as reported by Ridtitid et al.). In the series reported by Ruiz de Gordejuela et al., 6 patients required early re-operation and at follow-up 11 required revision. These inferior outcomes are noteworthy as they will have impacted significantly on patient care. The lack of success in endoscopic procedures will have resulted in repeat procedures with associated risks and in the case of outcomes reported by A. G. Ruiz de Gordejuela et al., the greater rate of revision surgery and complications will have worsened patient outcome. The greater rate of positive surgical margins should be emphasised as an area of concern as going forward this may adversely affect the oncological outcome for the patient.

These patient safety concerns have led some institutions to ban the use of LBSP in their meetings, such as the American College of Surgeons and the American College of Obstetricians and Gynaecologists. [3] Others such as the Royal Australian and New Zealand College of Ophthalmologists (RANZCO) have advised its use only in exceptional circumstances and have set out a checklist of points to be analysed prior to approval. In all other cases pre-recorded video is the preferred option.

It should be noted that a large proportion of LBSP remains unreported. Therefore, there is no rigorous analysis of patient outcomes and hence the level of patient safety remains unknown in these events. There may also be reluctance to report on events that show inferior patient outcomes. This, along with the availability of alternatives to LBSP, may have discouraged some surgical societies from adopting its use.

These studies highlighted a number of concerns with LBSP which can be of significant impact on patient care and outcome. This highlights the stressing need for formal regulation of LBSP and developing a quality assurance framework that encompass dedicated consenting, maintaining confidentiality as well as a reporting on operative and postoperative outcomes for patients involved in LBSP. Additional, further studies are required to develop objective evaluation of the potential educational values of these events and translate these benefits into a safer environment such as the use of pre-recorded unedited videos to minimise patient harm.

The potential negative impact of live surgery broadcast can be explained by a number of factors including, surgeons, fatigue, operating with unfamiliar teams and equipment in addition to the pressure to perform in front of a large audience.

Khan et al. explored LBSP from the view point of the surgeons performing the procedures. A small proportion of surgeons reported significant anxiety and reduced surgical quality when performing LBSP. This proportion increased significantly when LBSP was being performed in a foreign institution [35]. Finch et al. explored surgeons’ views on the educational value and safety of LBSP verses ‘As-live unedited surgical broadcast’ (ALB). The educational value was felt to be similar between both formats but it was felt there are greater patient safety benefits with ALB. Also of note respondents were significantly less likely to recommend themselves, a relative or friend to be a patient in LBSP [36].

In order to ensure safety at events with LBSP, surgical societies and institutions have published 14 guidelines and frameworks to maximise patient safety. While there are a small number of societies who outright ban LBSP at their meetings [3], a larger proportion note the educational value and advise operating surgeons and physicians to work in the scope of the guidelines in order to uphold patient safety as the priority.

Consent

This review has also compared the outcomes of the 13 original studies against the recommendations from the summary of 14 prior guidelines and position statements. Although there is unanimous agreement among all the published guidelines for the need to have a dedicated consent process for LBSP, this was not the case when the 13 papers were reviewed. This highlighted the need to improve the quality of reporting on the consent process during LBSP, stating the separate risks of being involved in live surgery, the potential added risks such as surgeon distraction and delays due to transmission.

Patient confidentiality

This review has also highlighted an important point related to protection of patients’ confidentiality during LBSP. This was clearly stated in all guidelines but not robustly reported on by the 13 studies. Standardisation of reporting on these important ethical matters in LBSP manuscripts should be an essential requirement to reassure the surgical community that these ethical standards of maintaining patient confidentiality are upheld throughout the LBSP.

Overall, there was a lack of recommendation on standardised approach of reporting on LBSP including the outcomes across the 14 published guidelines and positions statements. This needs to be a requirement in order to evaluate the safety of LBSP. This has been supported by the EAU, which has an established LBSP registry. This includes an application form as well as a check list which incorporates a wide range of requirements such pre-operative checks (selection of surgeon, consent and preparation of the patient), intraoperative (personnel, presence of moderators) and postoperative care (entering outcomes into the EAU live surgery registry, daily dedicated ward round and communication with the operating surgeon if any deviations occur).

This study has a number of limitations. We acknowledge that number of the studies in this review were modest with no randomised controlled trials and have moderate risk of bias in the quality assessment criteria. However, the 13 studies included a considerable sample size with some comparing LBSP to a matched controlled group. Secondly, we acknowledge that not all live surgical procedures broadcast are reported in the literature, which limits the generalisability of this review and the ability to capture key data to analyse the true educational value and safety of LBSP. Additionally, we conducted this review only on English-language studies and relevant articles in other languages may have been missed. Furthermore, most of the included studies did not report on fundamental issues such as measuring the potential educational values vs. the potential risk of LBSP. It was therefore not possible to draw a conclusion based on these reports. Finally, it was not possible in this synthesis to produce a meta-analysis due to the heterogeneity in reporting outcomes and complications among the included studies.

Nevertheless, this review has heightened a number of points in relation to the relevance of LBSP as part of virtual teaching curriculum during and post the pandemic. First, adherence to a high level of standards during LBSP must be maintained during and after the broadcast including maintaining patient confidentiality and reporting on patient outcomes. Secondly, there are logistical issues with conducting LBSP and the educational values must be balanced with those concerns including patient safety and confidentiality. Alternative teaching ways should also be explored such as streaming pre-recorded unedited videos, allowing the audience to see the procedure being performed while also gaining useful insights on how the surgeon deals with unexpected events, while minimising the risk by being outside the environment of a live surgical event. Finally, there is a need to study the role and potential impact of live demonstration on simulation models such as human cadavers, which could enhance the teaching experience without impacting on patient safety.

Conclusions

Live Broadcast of Surgical Procedures can be of educational value but patient safety and outcomes may be compromised. Further research is required to develop a standardised framework of reporting on live surgery and its outcomes from an ethical and patient safety perspective.