Audit Section

Prospective audit of quality of colonoscopy in a surgical coloproctology unit

J.S. Varma, T. Fasih and M.A. Tabaqchali
Coloproctology Unit, Department of Surgery, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP

Correspondence to: J.S. Varma, Coloproctology Unit, Department of Surgery, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP

Introduction Patients and methods Results

Discussion References

Objective: The aim of this study was to prospectively audit the quality of colonoscopy and patient acceptance in a Surgical Coloproctology Unit over a one-year period. Patients and Methods: 202 consecutive colonoscopies were evaluated over a 12-month period performed by a Consultant, Specialist Registrars and Research fellows. Data where recorded for adequacy of bowel preparation, completion rate, adequacy of sedation, patient tolerance and duration of the procedure. Adequacy of bowel preparation was monitored by scoring bowel content and the percentage of bowel wall visualised. Patients completed a questionnaire to express their sedation satisfaction, discomfort during the procedure and overall satisfaction. Results: The success rate of bowel preparation was 94%. Completion rate was 90% in intended full colonoscopies by the Consultant and Registrars and 74% by more junior grade endoscopists (overall 86%). The mean dose of midazolam and pethidine was higher in patients with unsatisfactory sedation than those with satisfactory sedation. The pain score was higher when trainees performed the procedure than when performed by the Consultant. Fourteen patients refused to undergo the procedure again due to procedure discomfort (n=7), inadequate sedation (n=2) and bowel preparation discomfort (n=5). Conclusion: A high completion rate was achieved, compared with published results. However, further improvements are possible especially by improving the performance of junior endoscopists and by ensuring optimal bowel preparation. Patients’ tolerance of colonoscopy was highly acceptable but may also be improved by the same methods

This paper was originally presented at the Annual Meeting of the Association of Coloproctology, UK & Ireland, Manchester, June 2002.

INTRODUCTION
Since the introduction of colonoscopy in 1960, the ability and quality of this technique has been enhanced by vast improvements in instrumentation enabling therapeutic intervention.¹ Colonoscopy continues to offer the highest sensitivity and specificity in diagnostic evaluation of the colon.² Successful colonoscopy depends on good bowel preparation, adequate patient sedation, insertion of the instrument to the caecum or terminal ileum and precise observation.³

Although reported completion rates range from 78% to 98%, competent colonoscopists should be able to intubate the caecum in at least 90% of cases.^4-8 This study was set up prospectively to evaluate the quality of colonoscopy in our unit over a one year period. The principle aims of the study were to assess completion rates, adequacy of sedation, bowel preparation and patient tolerance of the procedure.

PATIENTS AND METHODS
Two hundred and two consecutive patients were prospectively evaluated (93 males and 109 females) over a 12-month period. Their mean age was 57 years (range 14-90 years). All procedures were performed in a modern endoscopy unit using video endoscopes with comprehensive equipment and nursing backup. Twin rooms were available with the consultant performing in one of the rooms. Data where collected and entered onto an Access database on the same day by the dedicated endoscopy nurse.

Grade of endoscopists
Fifty-nine procedures were performed by the Consultant, 93 by the Specialist Registrars and 50 by less senior doctors, usually research fellows. 

None of the 93 endoscopies performed by Specialist Registrars were routinely supervised, although this was available ( twin rooms ). However, 16 (8%) of the endoscopies performed by the juniors were supervised. Juniors were encouraged to seek supervision if required.

Sedation
Conscious sedation was aimed for. Before the procedure, patients were made aware that they would receive sedation which would make them drowsy but not asleep because it was important that they could be communicated with easily throughout the procedure.

Intravenous pethidine and midazolam were administered immediately prior to the procedure in the endoscopy room. All 202 patients received sedation in the form of midazolam and 197 patients were given pethidine as well (97%).

Completion of procedure
The procedure was deemed to be complete when the caecum was reached. The ceacum was identified by any combination of visualisation of the opening of the appendix, ileocaecal valve, trivalvular folds and a positive transillumination in the right iliac fossa.

Duration of procedure 
The endoscopy nurse noted the duration of the whole procedure and the time to reach the caecum or the most proximal point.

Pain scoring
The same endoscopy nurse assessed the pain experienced by the patient during the procedure using the Hayward pain scale, as shown in Table 1.9

Bowel preparation
Two hundred and fifty patients underwent full bowel preparation. Patients were instructed to take clear fluids only on the day before the procedure.

Sodium picosulphate (Picolax) was used for complete colonic cleansing in a dose of one sachet orally in the morning and one in the evening of the day before the procedure. Patients were instructed to fast from midnight onwards and underwent the procedure the following morning.

The quality of bowel preparation was assessed by bowel content visualised (score 0-3) and the percentage of bowel wall visualised (score 0-3) in four parts of the large bowel i.e. the recto sigmoid, descending colon, transverse colon and right colon. The scoring system used to assess the success of preparation was a modified version previously used in another study and is shown in Table 2.⁶

Patient satisfaction
Patient satisfaction was evaluated by means of a questionnaire, completed by the patients more than 24 hours after the procedure, when the effect of the drugs had worn off. The two main determinants for evaluating the success of the procedure were satisfaction with sedation and discomfort during the procedure. Satisfaction with sedation was graded from extremely dissatisfying to extremely satisfying sedation, as shown in Table 3.

Patients experiencing pain during the procedure were asked to mark yes or no on the questionnaire to any grade of discomfort they experienced during the procedures using the pain scores described above.

Statistical analysis
Categorical responses were compared using the X2 test of independence. For quantitative numeric responses one-way ANOVAs were performed.

Figure 1: % Bowel wall visualised

RESULTS

Bowel preparation
The preparation was judged adequate in 187 patients (score= 0 - 2 for either bowel content or bowel wall visualised) but failed in 15(7.4%) patients (score >2 for either content or vision). (Figures 1 and 2). Overall, seven of these patients agreed to have a repeat full bowel preparation with a successful colonoscopy score of 1-2 for either bowel content or the bowel wall visualised.

Sedation
The mean doses of midazolam and pethidine were 6 mg and 50 mg, respectively. Sedation was judged to be satisfactory in 87% of patients (n=176) and unsatisfactory in 12.8%(n=26). The mean dose of midazolam was higher in patients in whom sedation was assessed as unsatisfactory, compared with those who had satisfactory sedation [p<0.01]. Similarly, the mean dose of pethidine was higher (54.4 mg) in patients with unsatisfactory sedation than those with satisfactory sedation (48.7 mg)[p=0.00]. Statistically, there was little difference in sedation satisfaction between the Consultant and the trainees (p=0.06).

Completion rates
Overall, the caecum was reached in 86% (n=174/202) of the procedures. Consultant and Specialist Registrars reached the caecum in 90% of the procedures. ’Others’ reached the caecum in 74% of the procedures. The various causes of failure to reach the caecum comprised obstructing tumours (n=4), stricture (n=1), patient intolerance (n=4), poor preparation (n=6), bowel perforation (n=1) and looping (n=12). The median duration to complete the procedure was significantly longer when sedation was judged to be inadequate (unsatisfactory 27 minutes vs. 22 minutes in the satisfactory group; Mann Whitney, p=0.04).

Figure 2: Bowel content visualised

Pain
There was a significant difference in mean pain score between the different grades of endoscopists as judged by the endoscopy nurse observer (p< 0.001). The mean pain score was 1.47 when performed by the Consultant, as compared with a mean pain score of two performed by the other trainees.

Patient satisfaction
One hundred and sixty-two patients returned the completed questionnaire.

Satisfaction with sedation: 14 patients were dissatisfied [8.6%(14/162)] with the procedure. Table 3 details the analysis.

Patient discomfort: Sixty-four patients (39%) experienced discomfort of varying degree, whereas 98 (60%) patients did not recall any discomfort during the procedure. There was no significant difference between the incidence of discomfort between the different grades of endoscopists (p=0.56).

Repeat test: Fourteen patients documented reluctance to undergo the procedure again. Their reasons were pain (n=7), inadequate sedation (n=2) and discomfort associated with the preparation (n=5).

DISCUSSION
Although colonoscopy has become the gold standard for diagnosis of colorectal pathology, it is not a perfect investigation.¹⁰ There are several factors affecting the success rate of colonoscopy, an important factor being bowel preparation.^3,4,6

Accurate colonoscopy requires adequate bowel preparation for a clear visualisation of the colorectal mucosa.^10,11 In our study, full bowel preparation was inadequate in 6% of the patients and caused repeat of the procedure in 2.8% of patients. In the majority of patients we used Picolax, therefore, avoiding larger volumes of preparation that can be unacceptable in many (5-15%) cases, especially the elderly.¹¹ Therefore, great emphasis should be given to proper bowel preparation for a successful procedure.³ In our series most of our incomplete examinations were due to poor bowel preparation as also reported by Church (1994).⁷

The administration of intravenous narcotics during colonoscopy has become standard practice. Patients in our study accepted the procedure better, as far as sedation was concerned, when the endoscopist was of a senior grade. This was probably related to better communication and technique. It is important that trainees learn to gently manipulate the scope rather than relying on more sedation. Colonoscopists should clearly understand the importance of intravenous narcotics as adequate sedation involves a conscious patient who can be communicated throughout the procedure. At the same time their use has been associated with a small but important risk for serious cardio-respiratory complications.¹²

Colonoscopy is considered complete when there is identification of the ileocaecal valve, the appendiceal orifice and the short caecal sling folds in the caput coli.⁴ Although ultimate definition of a “total colonoscopy,” is entry of the terminal ileum, this is not the usual or recommended practice. The overall results were significantly affected by failure to complete the procedure by juniors.This was due to the juniors being left to ask for help when necessary but failing to do so sufficiently. We have changed our practice to routine supervision now. A more supervised procedure would have had a better outcome in this series, although some studies have shown that trainees do not achieve competence even after a threshold of 100 supervised colonoscopies.⁶ The major part of colonoscopy teaching by trainers includes the presence, positive discussion and positive criticism.¹³

We conclude that, in our series, the quality of colonoscopy could have been improved by routinely supervising the juniors before they are allowed to perform procedures on their own, as also recommended by the American Society for Gastrointestinal Endoscopy. Adequate bowel preparation and patient counselling also plays a vital role. 

Our Unit presently provides an acceptable quality of colonoscopy as measured against the published standards, but, based on this audit, further improvements may be possible.

REFERENCES
1. Waye JD.’Colonoscopy my way’: preparation, anticoagulants, antibiotics and sedation. Canadian Journal of Gastroenterology 1999;13((6): 473-6.
2. Farrell RJ, Morrin MM, McGee JB.‘Virtual Colonoscopy: a gastroenterologist’s perspective’. Digestive diseases. 1999;17(4):185-93.
3. Won Ho Kim, Young Jun Cho, Jeon Y Park, Phil K Min. Factors affecting insertion time and patient discomfort during colonoscopy. Gastrointestinal End 2000 ;Vol 52, No.5; 600-605.
4. Waye JD. Editorials. What constitutes a Total Colonoscopy? The Am J of Gastroenterology 1999 Vol. 94.no.6: 1429-30.
5. Isbister WH. Colonoscopy: How far is enough? Aus New Zealand J Surg 1995; 65:44-7.
6. Chak A, Cooper GS, Blades EW, Cauto M, Sivak MV Prospective assessment of colonoscopic intubation skills in trainees. Gastroint. Endoscopy 1996; Vol.44no.1: 54-57.
7. Church JM. Complete colonoscopy: how often? And if not why not? Am J of Gastroenterology 1994; 89(4): 556-60.
8. Marshall JB, Barthel JS. The frequency of total colonoscopy and terminal ileal intubation in the 1990s. Gastrointestinal Endosc 1993; 39:518-20.
9. Hayward J. Can pain be measured? Nursing (London) 1979 April(1):32-34.
10. Haseman JH, Lemmel GT, Rahmani EH, Rex DK. ‘Failure of colonoscopy to detect colorectal cancer: evaluation of 47 cases in 20 hospitals’. Gastrointestinal Endoscopy 1997:Vol.45, No.6, 451-455.
11. Habr-Gama A, Bringel RWA, Nahas SC, Araujo SE, Souza Junior AF, Calache JE, Alves PA. ‘Bowel preparation for Colonoscopy: Comparison of Mannitol and Sodium Phosphate. Results of a Prospective Randomised Study’. Hosp. Clin. Fac. Med.S. Paulo 1999;54(6): 187-192.
12. Marrow J B, Zuccaro G, Conwell D L, Vargo J J, Dumot J A, Karafa M, Shay S S. ‘Sedation for Colonoscopy Using a Single Bolus is Safe, Effective and Efficient: A Prospective, Randomised, Double-Blind Trial’. The Am J of Gastroenterology 2000; Vol.95, No.9, 2243-47.
13. Teaguee RH. ‘Can we teach colonoscopic skills?’ Perspectives. Gastrointestinal Endoscopy 2000 Vol. 51.No.112-4.
14. Freeman B, Engel JJ, Fine MS, DiVita DP. Colonoscopy to the caecum: how often do we get there? Experience in a community hospital [letter]. American J of Gastroenterology 1993; 88(5).

Copyright: 16 February 2004

FORTHCOMING DENTAL FACULTY EVENTS

* MFDS/MFD (Parts A & B) Preparatory Courses 17th & 18th July 2004

* Dental Symposium “Dental Implants” 23rd September 2004

* Masterclass by Professor Takemoto “Lingual Orthodontics Typerdont” 28th & 29th November 2004

* Masterclass by Dr James Mah “Craniofacial Imaging” 30th November 2004

For further information please contact:

Mrs Violet Brown, Dental Faculty Administrator

The Faculty of Dental Surgery, Royal College of Surgeons of Edinburgh

10 Hill Square, Edinburgh, EH8 9DW

Telephone Number: 0131 527 1608 Fax Number: 0131 527 1669 E-mail: dental@rcsed.ac.uk