Introduction Following the quality improvement programme introduced to pave the way for the UK Bowel Cancer Screening Programme (BCSP), there was an improvement in caecal intubation rate from 76.9% in 1999 to 92.3% in 2011.1 90.2% suffered mild discomfort or less, although this was assessed and reported by the endoscopist. 53.6% of symptomatic patients had abnormal examinations but it was not clear how many of the pathologies identified were incidental findings. We aimed to determine the percentage of patients whose colonoscopy findings explained their symptoms (the diagnostic yield) and to compare endoscopist and patient assessment of colonoscopy tolerance.
Methods Endoscopists and patients undergoing colonoscopy independently provided a Gloucester comfort score (GCS 1: comfortable; 5: severe discomfort). Endoscopic and subsequent histological findings were recorded.
Results Symptomatic patients (n = 107; 46.7% male, mean age 55.0±1.6 yrs) were younger than BCSP patients (n = 46; 65.2% male, mean age 67.0±0.9; P < 0.001, two sample T test). In the symptomatic group, in whom the diagnostic yield was 17%, moderate to severe discomfort (GCS 4–5) was reported by endoscopist and patient in 21.7% and 46.2% (P = 0.001, Mann Whitney) and in the BCSP group, by 21.8% and 26.1% (P = 0.6) respectively. Moderate to severe patient discomfort was more common in the symptomatic than BCSP group (P = 0.019), but overall was less common in males than females (P < 0.02). Sedation had no effect on tolerance (P = 0.9).
Conclusion Almost half of patients having diagnostic colonoscopy reported moderate to severe discomfort, which was markedly underestimated by endoscopists. If the true diagnostic yield is only 17%, it would seem appropriate to offer patients a less invasive, better tolerated alternative as a first line test. Procedural tolerance was much better in the BCSP, where endoscopists were more accurate in grading patient discomfort, which may partly be explained by the male predominance. However, GCS 4–5 scores were still three times as common as reported in the national audit.
Reference 1 Gavin DR, et al. Gut 2013;62:242–249.
Disclosure of Interest None Declared