Article Text


Endoscopy free papers 140–149

Statistics from


C.J.A. Bowles1, R. Leicester2, E. Swarbrick3, C.B. Williams4, C. Romaya5, O. Epstein1. 1Royal Free Hospital, London; 2St George's Hospital, London; 3New Cross Hospital, Wolverhampton; 4St Mark's Hospital, London; 5Audit Office, B.S.G., London, UK

Introduction: Colonoscopy may be complicated by bleeding, perforation and sedation related side effects. The General Medical Council guidance on consent states that sufficient information should be provided on the purpose of a proposed investigation or treatment as well as common and serious side effects. Patients should be allowed sufficient time to reflect before making a decision. As part of the IBNC audit a patient questionnaire ascertained details of consent with regard to whether written information was provided, where consent was obtained and whether adverse effects were considered.

Results: 1200 patient questionnaires were distributed and 599 (49.9%) returned. Prior to colonoscopy 488/599 (81.5%) patients received written information. In 328/599 (54.8%), consent was obtained in the procedure room, and in 179/599 (29.9%), immediately prior to the colonoscopy but not in the procedure room. In 60/599 (10.0%), consent was obtained as an outpatient and 32/599 (5.3%) patients couldn't remember or didn't answer this question. Possible adverse events were reported by 329/599 (54.9%) patients and bleeding and perforation were specifically cited by 95/329 (28.9%) and 96/329 (29.2%) respectively. No mention of adverse events was reported by 196/599 (32.7%) of patients and 56/599 (9.3%) couldn't remember whether or not they were provided with information on possible adverse events. 18/599 (3.0%) provided no response to the question on adverse events.

Conclusion: The majority (81.5%) of patients were provided with written advice prior to colonoscopy. Contrary to the GMC advice, patients are frequently consented in the procedure room and only 54.9% of patients reported that they had been informed of adverse effects. In the interests of ensuring best practice, endoscopists should be constantly mindful of the GMC guidelines on consent.


S. Ramakrishnan, J.Y. Yiannakou, T. Butler, W.R. Ellis, I.M. Bain. Department of Gastroenterology, University Hospital of North Durham, Durham, UK

Background: Endoscopists focussed on the technical challenges of colonoscopy may not adequately appreciate patient discomfort.

Methods: We conducted a prospective study of 474 colonoscopic procedures most of which were performed by 5 endoscopists. Overall completion rate was 89 %. Pain was assessed during the procedure as visual analogue scores (0 to 10) by the nurse, endoscopist (doctor) and patient. All were blind to the others' scoring. Patients had variable amounts of sedation and analgesia. A Multivariate Linear Regression Analysis was performed on 426 data points.

Results: Data was complete on 426/ 474 questionnaires. The average score for doctor, nurse and patient was 2.8 (S.E 0.1), 3.09 (S.E 0.11) and 3.2 (S.E 0.13) respectively. Pain scores of doctors and nurses were compared to that of the patient. The correlation coefficient was 0.42 (p< 0.01) and 0.59 (p< 0.01) respectively, both highly significantly related to patients' perception of pain. Indeed nurses appeared to have a better perception of this pain. When using a multivariate analysis, modelling patient pain on both doctors and nurses perception of pain, the doctors have little predictive value over and above nurses, i.e. doctor's perception is no longer significant when adjusted for nurses' perception (p=0.39). However nurses' perception remains highly significant when adjusted for doctors' perception (p<0.01)

Conclusions: The degree of unpleasantness/pain recollected by patients is the most important factor in the acceptability of this procedure. Nurses were able to provide a closer assessment to this than the endoscopist (doctor). This maybe because endoscopists are focussed on the video monitor while nurses are focussed on the patient. This may suggest a need for better training of endoscopists or more active use of nurse's assessments during the procedure for achieving best results with minimal patient discomfort.


C. Mowat, M. McKernan1, J. Morris. Depts of Gastroenterology and 1Surgery, Glasgow Royal Infirmary

Introduction: Direct access endoscopy referral forms have been used by local GP's for 2 years. The local Dyspepsia Management Guideline recommends urgent referral for dyspeptic patients with dysphagia, weight loss, anaemia, recurrent vomiting. There is little evidence to support the value of these symptoms in predicting serious pathology. We reviewed the outcomes of patients referred with alarm symptoms in our population.

Methods: A 12 month retrospective study was performed (Sept 00 - Aug 01). Endoscopy lists, referral forms and outcomes were obtained from centralised records. Pathology records of all upper GI cancers presenting in study period were collected.

Results: 597 GP referrals were endoscoped. 274 (46%) had alarm symptoms; mean age 58 (19–87). Symptom frequencies were dysphagia 32%, wt loss 50%, anaemia 14%, vomiting 35%. In those with alarm symptoms, 88 had a normal endoscopy, 99 had some gastritis/duodenitis, 39 had oesophagitis alone, 20 had peptic ulcer, 5 (1.8%) gastric cancer, 7 (2.5%) oesophageal cancer, 9 (3%) had Barretts. None of the alarm symptoms were predictive of a particular diagnosis. Over the study period, a total of 67 upper GI cancers presented to the hospital (28 oesophageal, 32 gastric, 2 duodenal, 3 pancreatic, 2 incomplete data); mean age 73 (46–90). Only one patient was aged less than 55. 26 of these presented via other direct access routes. The remainder presented via A&E or clinic. 12 of 38 patients referred urgently versus 2 of 22 in-hospital referrals were suitable for attempted curative surgery. Of the 14 suitable for surgery, mean age was 68 (58–75).

Conclusions: In our population, alarm symptoms in young dyspeptic patients rarely indicated upper GI cancer. The cancers arose in older subjects. Those who were referred urgently had a significantly greater chance of attempted curative surgery. Rapid assessment of dyspeptic patients with alarm symptoms should therefore focus on older patients. In this study, all patients with alarm symptoms aged less than 55 could have been managed empirically with a `test and treat' +/- PPI policy.


C. Roseveare1, L. Bright1, D. Dalgleish1, J. Kimble, H. Shepherd. 1Southampton General Hospital; Royal Hants County Hospital, Winchester, UK

Background: Our previous studies have indicated that Patient-Controlled Sedation (PCS) using propofol and alfentanil provided an effective alternative to a combination of diazepam and pethidine given as a bolus prior to the procedure, with the advantage of shorter recovery time.

Aims: To compare efficacy of sedation and recovery time between PCS and a bolus combination of midazolam and pethidine (M&P); to study further the safety of the technique and to determine the feasibility of PCS being set-up and supervised by an endoscopy nurse and endoscopist.

Methods: 67 patients undergoing colonoscopy were randomised to receive sedation with either PCS using propofol and alfentanil or a bolus of midazolam (2.5–5mg) and pethidine (25–50mg). Infusions were connected by the endoscopy nurse with anaesthetist present in an observational capacity only. The anaesthetist only intervened if specific pre-defined criteria were reached. Sedation scores were recorded during the procedure by the endoscopy nurse and pain scores after the procedure by both nurse and patient, using likert scales. Recovery was confirmed using number connection tests. Impact on subsequent daily activity, amnesia and overall satisfaction were established by phone at 24 hours.

Results: Sedation method had no impact on the success, difficulty or duration of the colonoscopy procedure. PCS infusions could be set up by the endoscopy nurse without causing significant delay. Patients in the PCS group recovered significantly faster (mean 5mins vs 35mins, p<0.0001) and left the endoscopy department much sooner (40mins vs 75mins, p<0.0001). No differences in respiratory or haemodynamic observations were recorded between the two groups. Anaesthetist intervention was required for one patient sedated with M&P, but was not required for any patients sedated with PCS. Overall sedation level was lighter in the PCS group (score 3 vs 4, p=0.05) and verbal contact was lost with only 2 patients using PCS, compared to 9 using M&P. Patients in the PCS group reported significantly more pain (median pain score 1 vs 0, p=0.0005), which may have reflected greater amnesia in the M&P group, since nurse recordings of pain were identical for the two groups. PCS had less impact on the patients' activities in the 24 hours after the procedure (medican score 0 vs 1, p=0.01). Satisfaction scores were equal in the two groups.

Conclusions: PCS using propofol and alfentanil is acceptable to patients, provides faster recovery, and appears to be at least as safe as a bolus combination of midazolam and pethidine.


G.A. Paspatis1, M. Manolaraki2, G. Xirouchakis2, N. Papanikolaou1, I. Charoniti1, A. Gritzali2. Department of 1Gastroenterology and 2Anesthesiology, Benizelion General Hospital, Heraklion-Crete, Greece

Background and Aims: A recent study suggests the synergistic sedation with a low dose of midazolam combined with propofol for conscious sedation in colonoscopies in a selected group of patients under the age of 60 years. Our study sought to compare the safety and efficacy of the synergistic sedation with a low dose of midazolam combined with propofol versus the standard regimen of midazolam and pethidine for conscious sedation in colonoscopies in a group of patients that included a sufficient number of elderly patients with several comorbidities, representative of the patient profile seen in public general hospitals.

Methods: 120 consecutive patients undergoing colonoscopy were randomly assigned to 1 of 2 medication regimens. Patients in group A (n=64, 33 over 65 years) received intravenously (IV) 2mg (<70 kg b.w.) or 3mg (>70 kg b.w.) of midazolam and a median dose of 80mg propofol. Patients in group B (n=56, 22 over 65 years) received IV a median dose of 5mg midazolam and 75mg pethidine. The patient's comfort level was assessed by a four point scale 24 hours after the procedure. The time to recover from sedation was assessed at 5, 10 and 30 min after the procedure by using the Aldrete score.

Results: Multivariate stepwise logistic regression analysis revealed that among sex, age, duration of the test, ASA grade and the sort of sedation, the synergistic sedation with midazolam and propofol was the only factor associated with a higher level of patient comfort (χ2=5.5, p<0.05) and a quicker patient recovery time (χ2=24.5-p<0.01, χ2=51.7-p<0.01 and χ2=148.4-p<0.01 for aldrete in 10, 20 and 30 min respectively). Endoscopist's evaluation of patient sedation and cardiorespiratory parameters were similar in both groups across all age groups.

Conclusions: Our data suggest that the sedation with midazolam combined with propofol was superior to combination of midazolam and pethidine for colonoscopies as far as the patient comfort and recovery times are concerned.


S.G. Shah, J.C. Brooker, C.B. Williams, C. Thapar, N. Suzuki, B.P.Saunders.

Background: Variable-stiffness colonoscopes (VS scope) combine paediatric flexibility shaft characteristics for negotiation of the sigmoid colon with the ability to stiffen to prevent looping after straightening. Previous studies have shown a wide variation in efficacy of the stiffening mechanism, based on subjective assessments and differences in methodology. Two studies were conducted to assess the potential benefit of the stiffening device and its optimal use.

Methods: In study 1, the effect of routinely stiffening the straightened VS scope (Olympus CFQ240AL) in the mid-descending colon was determined in 82 patients. Two insertions were performed in each patient, from the mid-descending colon to caecal pole, with and without application of the stiffening device (randomised). The time taken to negotiate the proximal colon (from the mid-descending to caecal pole), time to pass the scope across the splenic flexure into the transverse colon, time to pass the right colon, and ancillary manoeuvres used were recorded for each insertion. In study 2, consecutive patients, excluding any with previous colonic resection, were examined using standard adult variable-stiffness colonoscopes. Real-time views of the procedure using magnetic endoscope imaging (MEI) were recorded in all examinations, but procedures were randomised to be done either with (n=88), or without the endoscopist viewing (n=87) the MEI display. Whenever shaft stiffening was applied, the anatomical location of the colonoscope tip and stiffness efficacy was recorded.

Results: Study 1—Time taken to negotiate the proximal colon (p=0.0041) and time to negotiate the splenic flexure (p=0.006) were significantly shorter and ancillary manoeuvres performed were fewer (p=0.0014), when insertion was carried out with the stiffening device activated. Study 2—Shaft stiffening was used with similar frequency in patients examined with and without the MEI view, most commonly for passing the splenic flexure (71%), but also in the transverse colon (12%), right colon (9%) and sigmoid/descending colon (8%). Shaft stiffening was significantly more effective when used in combination with MEI (69% with imager vs. 45% without imager; p=0.0102).

Conclusions: Overall, the variable-stiffness device was effective in controlling looping 57% of the time. Activating maximum scope stiffness appears to be most effective once the sigmoid colon has been negotiated and the scope straightened with the tip in the proximal colon, reducing the number of ancillary manoeuvres and shortening the insertion time through the proximal colon. Routine colonoscope imaging with MEI further enhances the efficacy of VS scopes by helping to identify the optimal time for scope stiffening.


S. Dogramadzi1, K. Burn2, R. Bicker3, G.D. Bell4. 1Department of Electrical and Electronic Engineering, Newcastle University; 2School of Computing, Engineering and Technology, University of Sunderland; 3Department of Mechanical Materials and Manufacturing Engineering, Newcastle University; 4Medical Sciences Faculty, Sunderland University

Introduction: The rather time consuming simple beam displacement method used to determine flexural rigidity (EI) along the length of an endoscope is reproducible and inexpensive (Gut 2001;49:154). To our knowledge all previous published results for the measurement of endoscope shaft stiffness have been performed at room temperature. We argued that measurements of shaft stiffness at body temperature might be equally important clinically (particularly during a prolonged procedure). We have developed an elegant computer-linked method of rapidly measuring EI such that the stiffness of the entire shaft of a typical 165 cm colonoscope can be determined in less than a minute.

Aims: To measure the EI along the shaft of a number of different commercially available endoscopes at both room temperature and again after 10 minutes immersion in a thermostatically controlled water bath at 40 degrees C.

Methods and Results: We measured EI in a range of different Olympus and Pentax colonoscopes and gastroscopes (n=12) that were in use on our Endoscopy Unit. In all cases there was a highly significant (p<0.001) 10–40% reduction in EI at 40 °C compared with room temperature.

Conclusions: Endoscopists need to appreciate that once an endoscope is inserted into a patient it will rapidly become significantly floppier as the endoscope warms up to body temperature. These observations might be particularly relevant to the problems related to recurrent looping that may be observed during a prolonged colonoscopic procedure.


S.G. Shah, M. Lockett, S. Thomas-Gibson, J.C. Brooker, M. Vance, C.J. Thapar, B.P. Saunders. Wolfson Unit, St Mark's Hospital, Harrow, London, UK

Background: Most trainees have little concept of the loops that occur during colonoscopy and have difficulty in appreciating the combination of withdrawal and torqueing manoeuvres, essential to achieving complete colonoscopy. Real-time magnetic endoscope imaging (MEI) allows visualisation of shaft looping, and so makes intuitive the manoeuvres necessary to straighten the colonoscope shaft.

Method: Consecutive routine colonoscopies were performed by a single trainee. Procedures were randomised to be carried out either with the trainee viewing the MEI display, or without the MEI view.

Results: In total 72 procedures were performed. To assess the trend for learning, procedures were analysed in blocks of 24 consecutive examinations (periods 1 to 3). See table.

Abstract 147

Conclusions: Real-time colonoscope imaging using MEI appears to enhance the endoscopist's appreciation of looping, and the learning of manoeuvres to straighten the colonoscope shaft, during training.


1J.C. Brooker, 1S.G. Shah, 1M. Vance, 2A.D. Millar, 3H.J. Pearson, 1B.P. Saunders. 1St Mark's Hospital, Harrow; 2North Middlesex Hospital, London; 3Diana Princess of Wales Hospital, Grimsby, UK

Background: The drawbacks of “hot-biopsy” include a small but important risk of complications and a 15–20% failure rate for adenoma eradication. Cold snare, an alternative technique, may fail to yield a specimen for histology. We hypothesised that simple biopsy followed by destruction with Argon Plasma Coagulation (APC) would be an effective and safe means for destroying small adenomas, whilst guaranteeing a histology specimen. This method was therefore compared with “hot-biopsy” in a randomised controlled trial.

Methods: Consecutive outpatients attending for flexible sigmoidoscopy were included if a suspected adenoma <5mm was detected. Subjects were randomly allocated to one of two groups, either conventional hot-biopsy or cold biopsy and APC. Standard diathermy settings were used (APC:65 Watts and 2L/min gas flow; hot-biopsy:15 Watts coagulating current). Hot-biopsy technique involved tenting the polyp and applying diathermy until visible mucosal whitening occurred at the base (the “Mount Fuji” effect). APC was applied to cauterise the entire polyp surface. A tattoo was placed adjacent to polypectomy sites using sterile india ink. Patients were contacted by telephone at 2 weeks to check for complications and those with confirmed adenomas were followed up with colonoscopy at 1–2 months. At colonoscopy the polypectomy sites were identified and biopsied to check for recurrence.

Results: From 505 examinations (237 male, median age 55[19–93]), 40 suitable adenomas (median size 4mm[2–5mm]; histology: 33 tubular, 7 tubulo-villous) were identified in 33 patients (19 males; median age 63[43–83]). 20 were treated with APC and 20 with “hot-biopsy”. There were no complications. At follow up colonoscopy, a median of 6.4 weeks later (range 3–11 weeks) there were two recurrent adenomas following “hot-biopsy”, but none after APC (p=0.49, Fisher's exact test).

Conclusions: These results indicate that argon plasma coagulation is at least as effective as “hot-biopsy” for destroying diminutive colorectal adenomas. We propose the development of a single accessory which combines biopsy forceps and Argon Plasma Coagulation together. This could offer a safe and efficient alternative to traditional “hot-biopsy”.


M. Mylonaki, A. Fritscher-Ravens, S. Lewkowicz, P. Swain. Royal London Hospital, Whitechapel, London E1 1BB, UK

Background: Wireless capsule endoscopy has the potential to deliver video images from the colon. Slow transit, intermittent rapid movements, larger diameter lumen and capsule transmission times are challenges requiring study.

Aim: To assess colonic images received during experimental, human volunteer and clinical studies. To improve the capacity of wireless capsule technology to image the human colon.

Methods: Colonic views obtained during clinical studies of capsules with a 7.5 hr capsule were reviewed (n= 38) human volunteers and patients referred with obscure gastrointestinal bleeding. Studies using experimental long-play with 4 silver-oxide batteries and a potential transmission time of 18 hours were used in 7. Lighter capsules [n=3](SG = 1.2 vs 1.7), with 2 batteries and a 2 hour rest period moving with the fluid phase were tested. 2 capsules were placed retrogradely into the transverse colon using hydraulic delivery.

Results: In clinical studies performed without colonic preparation (n=38) interpretable images of the human colon were acquired in 35, mucosal detail was seen in 34, faecal material 35, underillumination 10. Technical improvements included shorter capsules (27mm, previ-ously 33mm) now transmitting for 7–8 hours using two (previously three) batteries, and experimental capsules with an 12 hour life were been tested and occasionally imaged the toilet. Clinical studies in n=13 patients) revealed missed colon cancer/polyp (3), bleeding Meckel's diverticulum (1), ulcerative colitis (2), angiodysplasia (2). Small intestinal transit times were median (, range (1.2–7.6hrs). Oro-anal transit times were 0.5–8 days. Capsules used in patients with incomplete colonoscopies provided useful clinical information. In patients selected for capsule colonoscopy management was altered in 50%.

Conclusion: Wireless capsule colonoscopy is feasible and has already delivered valuable information in clinical studies. Technical development and better preparation and control over colonic motility is required to extend the range and improve the images of the colon in patients. Capsule colonoscopy was a useful adjunct to conventional colonoscopy especially in patients with difficult recurrent bleeding or incomplete colonoscopy.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.