Prevalence of surgical video recording in the United Kingdom and Ireland: results from the

Observational Study of Camera-Assisted surgery Recording (OSCAR) study

INTRODUCTION

This international, prospective study aims to

assess the prevalence of surgical video recording

in the United Kingdom and Ireland at individual

case-level. A previous assessment of surgical

video recording adoption at information

governance-level estimated 15.7% NHS trusts in

England/Wales routinely record surgical

procedures, though this did not provide

resolution about practice at individual case-level.

METHODS AND PROCEDURES

This multicentre, prospective study was

undertaken at 18 institutions in the UK and

Ireland, and included patients from 14 surgical

specialties. The primary outcome measure was

30-day postoperative mortality. Secondary

outcome measures were length of stay (LOS) and

30-day Clavien-Dindo grade of worst

complication.

RESULTS

1064 patients underwent a surgical procedure

between 11.18.24 and 2.23.25 with 30-day follow

up; 464 (43.6%) emergency and 599 (56.3%)

elective. Video recording took place in 65 (6.1%)

patients; approaches were open (12/65; 18.5%),

endoscopic (10/65; 15.4%), laparoscopic (17/65;

26.2%), and robotic (26/65; 40.0%).

Surgical subspecialties were colorectal (29/65;

44.6%), upper GI (17/65; 26.2%), urology (2/65;

3.1%), neurosurgery (17/65; 26.2%).

Differences in 30-day mortality were not

statistically significant (p=1.000); 0.0% (0/65) in

the video group and 2.8% (28/999) in the non-

video group.

Differences in mean LOS were not statistically

significant (p=1.000); 4 days (IQR 2-7) in the video

group and 1 day (IQR 0-3) in the non-video group.

Differences in Clavien-Dindo grade III or above

complication rate were not statistically significant

(p=1.000); 6.15% (4/65) in the video group and

3.7% (37/999) in the non-video group.

Intuitive Hub was the commonest recording

platform, used for all 26 recorded robotic

procedures (26/65; 40.0%), followed by Karl

Storz AIDA which recorded 18 procedures via

other approaches (18/65; 27.7%).

In the video group, video was reviewed in 32.3%

(21/65) cases; viewing purposes were stated as

quality assurance (0), education (11), research (11),

other (0).

Video review did not take place in cases where

there was a Clavien-Dindo grade III or above

complication.

DISCUSSION

The lack of easily demonstrable immediate

patient or surgeon benefit may hinder

behaviour change towards routine recording, in

turn preventing later benefit from a “digital

surgery dividend” of subsequently improved

surgical performance through case review and

other performance analytics.

Although previous studies have demonstrated

value in surgical video data for documentation,

quality assurance, education, training, and

research, a 1.97% (21/1064) surgical video

utilisation rate suggests that this value is not

being translated into the vast majority of patient’s

care within these health systems.

CONCLUSION

This is the first international study assessing

surgeon-level surgical video recording practice in

the UK and Ireland. Surgical video recording rates

are low, quantifying the challenge in generating

diverse surgical video datasets at scale for quality

assurance and building foundational data sets for

AI discovery. These data suggest video recording

is not associated with inferior patient outcomes,

but prospective trials are needed to establish

patient benefit from routine surgical video

recording.

Andrew Yiu, Kyle Lam, Catherine Simister, James Kinross on behalf of the Digital Surgery Collaborative

Imperial College London