Introduction Endoscopic submucosal dissection (ESD) is superior to endoscopic mucosal resection(EMR) as it leads to en-bloc resection & reduces risk of recurrence. However, western uptake has been limited due to long learning curve and procedure time.We aim to establish the time required for each component of ESD and identify factors predicting it.

Methods A single, experienced, western Endoscopist performed all procedures for suspected Barrett’s cancers >2 cm. Procedures were recorded in full on a digital recorder. First 30 procedures were considered as a part of learning curve and not analysed. All subsequent consecutive procedures were analysed by an independent researcher with knowledge of ESD.Lesion area was calculated based on length of the lesion and the percentage of the oesophageal circumference involved. Using the equation area = 2*π* r * l *% circumferential involvement / 100. Here r = radius and l = length of lesion. The time for every component of the procedure was recorded: lesion evaluation & marking, submucosal (SM) injection, mucosal incison, SM dissection, haemostasis & post-ESD site evaluation.

Results 29 consecutive videos were examined. All lesions were Barrett’s cancers (25% T-1b, 75% T-1a). The mean length was 30 mm (range:20-70mm), with mean area of 8.2cm² (range:1.6-23cm²). The mean procedure time was 81 mins (range:45-142min), equating to 9.9 min/ cm² .The time taken for each component of the procedure is shown in fig. 1. Only 42% of the time was spent in cutting (Mucosal incision and SM dissection). 24% of time was spent in evaluation and marking the margins. 24% of the time spent in changing accessories & injection. Procedure time was related to lesion area: 100 min for lesions >10cm² vs 72 mins for lesions <5 cm²(p = 0.0056).Circumferential extension had an effect, with <25% circumference taking 66 min vs 92 mins for lesions with >25% circumferential extension (p = 0.0025)

Conclusion Our data shows that it takes 9.9 min/ cm² to perform ESD for Barrett’s cancers. Time taken is directly related to the size and circumferential extent of the lesion. We found that only 42% of the time is spent performing the actual resection and rest of the time is spent in supporting acts. This information can help focus the future research in reducing the ESD procedure time and also help plan appropriate time and remuneration for current ESD procedures.

Disclosure of Interest None Declared