Quantitative and individualized assessment of the learning curve using LC-CUSUM.

BACKGROUND: Current methods available for assessing the learning curve, such as a predefined number of procedures or direct observation by a tutor, are unsatisfactory. A new tool, the cumulative summation test for learning curve (LC-CUSUM), has been developed that allows quantitative and individual assessment of the learning curve. METHODS: Some 532 endoscopic retrograde cholangiopancreatographies (ERCPs) performed by one endoscopist over 8 years were analysed retrospectively using LC-CUSUM to assess the learning curve. The procedure was new to the endoscopist and monitored prospectively in the initial study. Success of the procedure was defined as cannulation and proper visualization of the duct(s) selected before the examination. RESULTS: Fifty ERCPs were considered unsuccessful. There was a gradual improvement in performance over time from a success rate of 82.0 per cent for the first 100 procedures to 96.1 per cent for the last 129 procedures. The LC-CUSUM signalled at the 79th procedure, indicating that sufficient evidence had accumulated to prove that the endoscopist was competent. CONCLUSION: LC-CUSUM allows quantitative monitoring of individual performance during the learning process.

ERCP: a review of technical competency and workload in a small unit.

BACKGROUND: ERCP is increasingly performed not only in large referral centers but also in smaller units. We sought to analyze the success rates of selective cannulation and intervention using the cumulative sum method and to document the workload in a small unit. METHODS: Indications, results, and interventions performed by one endoscopist were recorded for all patients undergoing ERCP at Dunedin Hospital. Selective cannulation and successful intervention were used as outcome measures and, using the cumulative sum method, compared to a target value of 90%. RESULTS: Over an 8-year period, 532 ERCPs were performed. Overall 91% and 81% of selective cannulation and interventions respectively, were successful. The cumulative sum method plot shows that satisfactory outcomes for selective cannulation were obtained after some 100 to 120 procedures and after some 120 interventions. ERCP was normal in 171 (32%) patients, stones were found in 169 (32%), and strictures in 81 (15%) patients. CONCLUSIONS: The cumulative sum method is a valuable tool to compare individual performance with a nominated target value and to ensure that an acceptable outcome is achieved and maintained. These results show that small units can develop and maintain expertise in ERCP if procedures are performed regularly.

Flexible sigmoidoscopy or colonoscopy for colorectal screening: a randomized trial of performance and acceptability.

In a randomized clinical trail to assess acceptability, yields, costs, and unwanted effects of screening procedures, 232 subjects (137 with family history of colorectal carcinoma or adenoma, 95 without) were offered either flexible sigmoidoscopy or colonoscopy. Subjects with polyps found on sigmoidoscopy were followed up by colonoscopy. The two procedures were similar in compliance (65%) and yield (19% adenoma, 15% hyperplastic polyps). Polyps of either type were more common in those with a family history (prevalence: 41% compared with 24% without family history, p = 0.04). Costs per procedure were 60% lower for sigmoidoscopy, but follow-up colonoscopy reduced this cost advantage to 20% per subject. The subjects found the preparation for sigmoidoscopy easier, but the procedure more uncomfortable and embarrassing, as colonoscopy was performed under sedation. In this hospital-based study, colonoscopy was as acceptable to subjects, and only slightly more costly than sigmoidoscopy. Advantages of sigmoidoscopy would be greater for use outside hospitals and with less intensive follow up.

Mycobacterium paratuberculosis and inflammatory bowel disease: frequency distribution in serial colonoscopic biopsies using the polymerase chain reaction.

An association between Mycobacterium paratuberculosis and Crohn's disease is suspected but the evidence remains controversial. Using a one-step DNA extraction procedure with the thermophilic protease PRETAQ and amplification by the polymerase chain reaction, M. paratuberculosis DNA was detected in 22% of patients with Crohn's disease, and in 13% of patients with ulcerative colitis. M. paratuberculosis DNA was not found in any biopsy tissue from control non-inflammatory bowel disease patients. The biopsy tissues in which M. paratuberculosis was detected all came from regions which were inflamed when viewed microscopically. Overall, 7.7% of biopsies from such inflamed areas were positive. This low frequency of detection could be explained on the basis of extremely low abundance of the organism in relation to the area of mucosa sampled, or be consistent with a non-aetiological role for M. paratuberculosis in inflammatory bowel disease.

Measurements of unsaturated vitamin B12-binding capacity and myeloperoxidase as indices of severity of acute inflammation in serial colonoscopy biopsy specimens from patients with inflammatory bowel disease.

Measurements of tissue content of myeloperoxidase, a constituent of neutrophil azurophil granules and of unsaturated vitamin B12-binding protein from neutrophil-specific granules, have been used to assess intestinal inflammation. This paper reports results of a prospective evaluation of such measurements in serial colonoscopy biopsy specimens from patients with inflammatory bowel disease. Histologic grading of acute inflammation was based on perceived numbers of neutrophil polymorphs in sections from an immediately adjacent biopsy specimen. The mean + 2 SD range for unsaturated vitamin B12-binding protein activity in homogenates of histologically normal specimens was 62 pg mg protein-1. Values increased progressively up to 900 pg mg-1 protein in the most severely inflamed specimens. Unsaturated vitamin B12-binding protein measurements generally distinguished among histologic grades of inflammation, whereas myeloperoxidase activities failed to do this, probably because substantial myeloperoxidase activity was found in uninflamed colonic mucosa, suggesting a non-neutrophil source for this enzyme.

Enzymes degrading bacterial chemotactic F-met peptides in human ileal and colonic mucosa.

Bacterial chemotactic F-met peptides have been identified in culture supernatants of intestinal bacteria and in human faecal dialysates. These potent inflammatory agents could play a role in intestinal inflammatory disorders should they cross the epithelial barrier of the gut. We have identified mucosal peptidases which degrade F-met-leu-phe (FMLP) in ileal and colonic mucosal biopsies obtained at colonoscopy. A carboxypeptidase, inhibited by D-L-benzyl succinate (BzS), accounted for more than 60% of total FMLP-ase activity, other uncharacterized peptidases contributing the rest of the activity against the intact peptide. An F-met deformylase, inactive against di- and tri-peptides, cleaves released F-met completing the degradation. Total FMLP-ase, carboxypeptidase and F-met deformylase activities were measured in serial mucosal biopsies from 15 control patients undergoing colonoscopy for occult bleeding with negative findings and from 15 patients with ulcerative colitis (UC) and 10 with Crohn's disease (CD). Highest activities were found in terminal ileum and lowest in the rectum. Total FMLP-ase and carboxypeptidase activities were similar in controls and UC patients but were substantially reduced in CD, especially in the terminal ileum (controls 493 +/- 146 and 116 +/- 73 nmol/100 micrograms protein per h, respectively and CD 231 +/- 96 and 41 +/- 36 nmol/100 micrograms protein per h, respectively (P = 0.0018 and 0.015). F-met deformylase activities were similar in all groups. There was no correlation between enzyme activity and severity of inflammation. FMLP degrading peptidases probably contribute to the mucosal barrier of the gut in regions of high bacterial colonization, limiting intestinal absorption and inflammatory responses to these potent bacterial products in the intestinal lumen.