How we can measure quality in colonoscopy?

Abstract

Measuring quality is a current need of medical services either to assess their cost-effectiveness or to identify discrepancies requiring refinement. With the advent of bowel cancer screening and increasing patient awareness of bowel symptoms, there has been an unprecedented increase in demand for colonoscopy. Consequently, there is an expanding open-discussion on missed rates of cancer or precancerous polyps during diagnostic/screening colonoscopy and on the rate of adverse events related to therapeutic colonoscopy. Delivering a quality colonoscopy service is therefore a healthcare priority. Colonoscopy is a multi-step process and therefore assessment of all aspects of the procedure must be addressed. Quality in colonoscopy refers to a combination of many patient-centered technical and non-technical skills and knowledge aiming to patient’s safety and satisfaction through a continuous effort for improvement. The benefits of this endless process are hiding behind small details which can eventually make the difference in colonoscopy. Identifying specific quality metrics help to define and shape an optimal service and forms a secure basis of improvement. Τhis paper does not aim to give technical details on how to perform colonoscopy but to summarize what to measure and when, in accordance with the current identified quality indicators and standards for colonoscopy.

Key Words: Colonoscopy, Quality assurance, Metrics, Standards, Outcome

Core tip: With the advent of bowel cancer screening and increasing patient awareness of bowel symptoms, there has been an unprecedented increase in demand for colonoscopy. Delivering a quality colonoscopy service is therefore a healthcare priority. Colonoscopy is a multi-step process and therefore assessment of all aspects of the procedure must be addressed. Quality in colonoscopy refers to a combination of many patient-centered technical and non-technical skills. Identifying specific quality metrics help to define and shape an optimal service.

INTRODUCTION

Colonoscopy is the cornerstone in diagnosis and management of colorectal disease allowing direct optical diagnosis, tissue sampling for histological analysis and therapy of colonic lesions[1]. Quality of colonoscopy practice is highly variable and there is increasing public awareness of missed cancers, incomplete procedures and of adverse events related to colonoscopy which are potentially preventable[2,3]. The establishment of important, measurable quality indicators (metrics) and minimum quality standards is essential to define and shape a quality colonoscopy service.

The current quality indicators and standards for colonoscopy are based on varying levels of evidence, ranging from local perceptions and expert consensus to evidence from randomized controlled trials. The terms “auditable outcome” (an important indicator for which no clear evidence base exists) and “quality standard” (an auditable outcome for which there is an evidence base that can support a minimum standard) have been introduced to help define quality in endoscopy[4]. Τhis paper does not aim to give technical details on how to perform colonoscopy but rather summarizes what to measure and when, in accordance with the current identified quality indicators and standards for colonoscopy.

HOW WE CAN MEASURE QUALITY

A colonoscopy service can be broken down into three main steps: pre-procedure, on the day of procedure and post-procedure (Figure 1). A high quality colonoscopy service should be patient-centered, evidence-based, cost-effective and adhering to best practice. Quality indicators and standards for each step of the colonoscopy service should be as simple and easy to audit as possible (Table 1).

Table 1 Quality metrics for colonoscopy as proposed by ESGE’s guidelines and BCSP in United Kingdom.

When to measure	Outcome to measure	Standard
Pre-procedure	Appropriateness	100% indicated
	Pre-assessment–bowel prep to use	100% of cases
	Patient information	100% of cases
	Awaiting time when positive test	< 4 wk (< 2 wk desirable)
On the day of procedure	Review of comorbidities, check of vital signs	100% on admission
	Informed consent	100% signed
	Decontamination of endoscopes	100% agreement with local policy
	Appropriate function and availability of endoscopes/equipment	100% checked by competent staff
	Equipment for resuscitation and monitoring	100% regular checks
	CO2 insufflation	100% availability
	CIR	> 90% unadjusted
	Use of reversal agents	< 1/500 cases
	Bowel cleansing	good/excellent > 90%
	Patient comfort	NA
	Polyp detection rate	Dependent on case mix
	Polyp retrieval rate	> 90%
	Time of scope withdrawal	> 6 min
	Complication rates	Bleeding < 1/100
		Perforation < 1/1000 (diagnostic)
		< 1/500 (therapeutic)
	Electronically based endoscopy report	100% attached to histology request
	Aftercare plan	100% provided at recovery area
	Time for recovery	NA
Post-procedure	Annual number of procedures/endoscopist	> 150 ( > 300 desirable)
	Adenoma detection rate	> 15% unadjusted to race or gender
	Time of histopathology report	< 15 d post-colonoscopy
	Patient feedback/delayed AEs	100% at 30 d
	Endoscopic Surveillance needed	100% agreement with guidelines

CIR: Caecal intubation rate; AEs: Adverse events; NA: Not available.

Figure 1 The cascade of colonoscopy. AE: Adverse event; ADR: Adenoma detection rate.

Pre-procedure

An appropriate indication for colonoscopy should be determined in 100% of cases. Guidelines for indications and contraindications for colonoscopy should be used as a filter to avoid unnecessary and potentially hazardous procedures[5,6]. Time-scheduling should be based on priority (surveillance vs symptoms suggestive of CRC) and urgent referrals should be seen more rapidly. In our opinion a 6-wk time limit should be the maximum waiting time for a routine colonoscopy and ≥ 85% of individuals initially offered a colonoscopy should finally undergo a colonoscopy[4].

We recommend nurse-led patient pre-assessment either in a dedicated clinic or by telephone consultation especially when this has not been done by the vetting gastroenterologist. The endoscopist needs to have complete information of patient’s medical history prior to colonoscopy; comorbidities such as clotting disorders, use of anticoagulants or anti-platelet agents, diabetes, allergies, renal function impairment, glaucoma, heart failure and factors related to the risk of endocarditis should be recognised prior to colonoscopy and instructions given to each patient should be driven by current recommendations and local policy[7-10]. The American Society of Anaesthesiologists (ASA) status and factors which could increase the risk and technical difficulty of colonoscopy, such as previous abdominal surgery (i.e., hysterectomy) or diverticular disease should be recorded[7,11].

Patient information leaflets should be available and sent out to patients as a routine, along with a copy of the consent form. Patients must be aware of why the procedure is being organised, what is involved and of the risks related to colonoscopy. They should be informed about the options for sedation in advance and the associated restrictions on travelling home[7].

A clean bowel is a prerequisite for a reliable and efficient examination[12,13]. Clear patient information, reduced fiber diet, regardless of type of bowel preparation used, help to maximise bowel cleansing[14]. PEG-electrolyte is the preparation of choice in patients with renal impairment although it does not eliminate the risk of acute renal failure and it is considered safer for patients with cardiac failure[15,16]. Adequate hydration is vital to protect against adverse events of bowel preparation while timing and in particular PM/AM splitting of administration of the recommended dose and assurance of patient’s understanding of the process also appear to be important[14,17]. Table 2 outlines patients at risk of electrolyte imbalance and documents those who of when should have an assessment of renal function prior to bowel preparation. Those with established renal disease, stage III or greater, should have PEG-electrolyte bowel preparation[18-22]. In our institution we use a combination of 10 senna tablets and 2 doses of sodium picosulfate the day before colonoscopy for morning appointments, while the second dose of sodium picosulfate is taken in the morning of the same day for afternoon colonoscopies. The patient is encouraged to drink at least 2 L of clear fluids daily for 2 d before the procedure and to avoid fiber 2 d before scheduled colonoscopy. We usually use a 2lt PEG solution (MOVIPREP) when needed. Although hospitalisation has been related with poorer bowel cleansing and should be routinely avoided, hospital admission prior to colonoscopy may be required in some cases, especially for patients in whom reduced absorption of regular medications may prove problematic and may need intravenous administration. Fragile patients with multiple comorbidities which are at risk of cardiac or renal failure and should be monitored during bowel prep are often admitted to hospital prior to colonoscopy[23]. Selection of these patients is a matter of careful clinical pre-assesment.

Table 2 Groups of patients in whom polyethylene glycol bowel-preparation is considered as safer and thus should be preferred.

Candidates for polyethylene glycol bowel preparation for colonoscopy
1GFR < 60 mL/min per 1.73 m2
Electrolyte imbalance
Cardiac failure
Liver cirrhosis
Hypertension with arteriosclerosis
Patients on diuretics
(when cannot be stopped 24 h prior to colonoscopy)
Patients on ACE inhibitors
(when cannot be stopped 72 h prior to colonoscopy)
Patients on NSAIDs
(when cannot be stopped 72 h prior to colonoscopy)

¹Estimated glomular filtration rate (GFR) from serum creatinine concentration. NSAIDs: Nonsteroidal antiinflammatory drugs; ACE: Angiotensin-converting enzyme.

Colonoscopy in obese patients may prove technically demanding in some cases however, in our practice and according to previous reports, routine colonoscopy is the screening test of choice and can be performed adequately in obese patients when optimal standards are fulfilled[23]. Patients with previous incomplete procedures, multiple comorbidities or on anticoagulant treatment in whom discontinuation can prove catastrophic should be offered a virtual colonoscopy (CT colonography) as an alternative. In these cases virtual colonoscopy may prove an important pre-assessment tool regarding the cost, tolerability and reduced time of the procedure compared with conventional colonoscopy[24,25].

On the day of the procedure

A brief review of the cardiorespiratory function including blood pressure, pulse rate and oxygen saturation in addition to documentation of adverse events related to bowel preparation or any medication given prior to colonoscopy (i.e., antibiotic prophylaxis) should be performed on the day of the procedure and before the patient’s entrance into the endoscopy room.

A signed informed consent should be obtained by 100% of patients prior to colonoscopy, ideally in a separate area rather than the endoscopy room where a patient’s privacy can be assured. Consent for colonoscopy must include a clear and realistic explanation of the procedure, possible attendant discomfort, the benefits and a clear discussion of risks and potential adverse events including sedation reactions, bleeding (immediate and delayed), perforation and missed pathology. Patient’s right to withdraw consent at any stage of the colonoscopy process should be understood by all members of the team[4,26]. Some institutions having the patient consented in clinic by the requesting consultant as well as giving the prescription for bowel preparation and patient leaflets and thus alleviating the need for postal issue for the same. This practice can prove beneficial acting as an indirect vetting as well of high risk patients.

Endoscopy room

The appropriateness, availability and functionality of the endoscopy room and equipment used during colonoscopy (including equipment used for patient monitoring) should be ensured through regular checks. Cleansing and decontamination of endoscopes should conform to current National or International guidelines[27].

Monitoring of vital signs (blood pressure, pulse and oxygen saturation) and regular checks of patient’s comfort and ability for verbal communication should be routinely used during colonoscopy. The use of CO₂ capnography is recommended to identify hypoventilation and hypoxia if heavy sedation required[28].

Patient’s comfort during colonoscopy is a critical quality outcome which refers to public acceptance rate of the procedure as a screening tool[29]. Levels of patient discomfort (no or minimal, mild, moderate, severe) should be recorded during colonoscopy.

The use of CO₂ insufflation, instead of air, is currently a quality standard to maximize comfort during unsedated colonoscopy and flexible sigmoidoscopy and permits reliable radiologic examination at the same day following colonoscopy[7,30]. Moreover, since carbon dioxide is an inert gas that cannot form a combustible mixture with hydrogen and methane, CO₂ insufflation avoids the very rare risk of explosion during colonoscopy with electrocautery and reduces post-polypectomy admissions after removal of large polyps[31,32]. Insufflation of CO₂ should be avoided in patients with COPD, known CO₂ retention or severely reduced pulmonary function.

The use of sedation improves patient tolerance of colonoscopy. A “titrated” (administrated gradually during procedure) low dose of an anxiolytic, such as midazolam (1.25-5 mg), given alone or combined with an opiate like pethidine (12.5-100 mg) or fentanyl (25-100 μg) are usually sufficient to achieve conscious sedation during colonoscopy[33], however, thresholds of pain and over-sedation remain undistinguishable and variable between individuals. Dosage reduction should be considered for older patients (> 70)[33-35]. Nitrous oxide/oxygen inhalation (Entonox) should be an alternative for people that cannot have intravenous sedation[36]. The type and dose of medications used the level of sedation (minimal-anxiolysis, moderate-conscious, deep or general anaesthesia) and the use of reversal drugs should be recorded at every colonoscopy and should be an auditable safety outcome.

The adequacy of colonic cleansing is an important outcome related to the reliability and completion rates of colonoscopy and should be reported at each procedure. Valid scales for assessment of quality bowel preparation have been made according to the presence of solid or semisolid stool and the relative limitation to achieving adequate visualization[37,38]. Excellent or adequate bowel preparation documented in > 90% of cases has been considered as a standard of bowel preparation efficacy[4,7].

Intubation of the most proximal part of the colon is a prerequisite to achieving complete examination. Intubation of the terminal ileum (TI) is not required if there is not specific indication while obtaining biopsies from normal TI is discouraged secondary to the relative concern of variant Creutzfeldt - Jakob disease’s transmission[39]. Caecal intubation rate (CIR) is a key quality indicator that reflects the performance skills of each colonoscopist, but can be affected by a variety of factors that can make the insertion of the scope difficult or impossible[40]. The main conflict in measuring the CIR of each colonoscopist is whether it should be adjusted for bowel preparation, obstructive lesions or for symptomatic patients. Overall, an unadjusted CIR > 90% can be used as the quality standard of colonoscopy, regardless of case[7].

The routine use of photodocumentation or videorecording is an emerging necessity in relation to the medicolegal risks of missed pathology or adverse events (AEs) following colonoscopy[41]. Photographic evidence of the appendix orifice and/or the ileocaecal valve has been considered as a standard practice to achieve completion[7]. Unarguably, additional pictures of the ileal mucosa provide strong evidence of completion[42]. Rectal retroversion has been considered as an established diagnostic technique to improve detection of lesions abutting the dental line[43,44] however an adequate examination can also be performed by tip manipulation in the forward view.

The incidence of colorectal cancer (CRC) can be significantly reduced through detection and appropriate removal of adenomatous polyps during colonoscopy[1]. The polyp detection rate (PDR) is defined as the number of colonoscopies at which one or more polyps were found (regardless of histological type) divided by the total number of colonoscopies performed (in the same time period). Counting polyps or polypectomy rates is easy during colonoscopy but is not as important parameter as adenoma detection rate (see later). A high retrieval rate (> 90%) of polyps removed is a recognized quality standard in the United Kingdom BCS program and can be affected by polyp size and cold snare technique of polypectomy[45]. The number and size of adenomatous polyps removed at colonoscopy should be recorded as this defines the risk of CRC and determines endoscopic surveillance[4,46,47].

Time spent on withdrawal (WT) is an important quality outcome and should be recorded during colonoscopy. A time for scope withdrawal of more than 6 min has been well-correlated with increased detection of adenomas and thus is considered as an important quality standard to be followed by each endoscopist[48]. Longer WT has been related with increased detection of proximal and serrated polyps[49,50]. Probably adequate withdraw technique and high technical endoscopist’s skills are more important to increase detection rate when appropriate WT (> 6 min) has been spent, but this is a matter of proper training and accreditation in colonoscopy that exceeds the purposes of this paper[51,52].

AEs in colonoscopy are uncommon but can be life threatening. Appropriate documentation of AEs related to colonoscopy is a substantial outcome of safety of the procedure. A Lexicon has been previously developed to provide clear definitions for AEs and levels of severity, including the minimum threshold at which an AE should be documented and reported[53]. Early AEs (bleeding, perforation, oversedation, vasovagal attacks), whether they have been adequately resolved during the procedure (i.e., use of haemostatic equipment or reversal drugs, hydration) or whether further actions are required, have to be clearly documented.

The endoscopist should be competent with the function of all supplementary equipment used during the procedure. Therapeutic colonoscopists should be technically competent to identify and safely remove high-risk lesions and be comfortable with techniques of endoscopic haemostasis[54,55]. Around 90% of post-polypectomy bleeding should be amenable to conservative management without the need for surgical intervention. According to current recommendations based on data from retrospective studies, the incidence of bleeding for colonoscopies where polypectomy is performed should not exceed 1/100[4]. However, this is a cut -off point that needs to be adjusted according to the time (immediate or delayed) and severity of bleeding, patients’ comorbidities and complexity of the procedure (i.e., EMR or simple polypectomy). Future analysis of risk factors for delayed bleeding should be possible and would optimally permit individualization of the risk of bleeding between patients. Risk of perforation should not exceed 1/1000 procedures, but may have to be adjusted to 1/500 for therapeutic colonoscopies with polypectomy[4]. In cases of therapeutic colonoscopy, the final report should include a clear description of “alarm post procedural symptoms” symptoms such as rectal bleeding, fever or abdominal pain that can be associated with delayed AEs requiring immediate medical support[4,56,57].

An increased number of AEs (ie bleeding or perforation) during therapeutic procedures always raise issues about the adequacy of therapeutic skills of each endoscopist. The European guidelines for quality assurance in colorectal cancer screening and diagnosis have proposed 5 levels of competency in colonoscopy related to the interventional armamentarium of each colonoscopist. According to this consensus colonoscopists should be able at least to remove lesions < 10 mm in order to avoid additional endoscopic procedures. We recommend that basic EMR technique for sessile polyps 1-2 cm in size, or for small flat adenomas smaller than 1 cm, should be within the armamentarium of all colonoscopists.

Recovery area

Standard protocols for monitoring and for emergencies should be available in the recovery area. Checks of availability and proper function of resuscitation and monitoring equipment should be regularly updated. Time of recovery is an important auditable outcome and should be recorded. After recovering from sedation and before leaving the endoscopy unit, patients need to be told about the outcome of their procedure in a simple and comprehensive way. Breaking bad news regarding suspicion of cancer should be done according to the established local policy. The average waiting time for the histopathology report and the aftercare plan should be provided and supported by a detailed written report of the procedure that includes a contact telephone number (24 h/d, 7 d/wk) in case of a procedure-related complication. An electronically based and formalized endoscopy report is essential for further interpretation of outcomes.

A copy of the endoscopy report should be attached to any histology request and should be as detailed as possible to provide accurate description of suspicious lesions including their location, their estimated size, their nature according to accredited classification systems (i.e., Paris or Lateral Spreading Tumors - LST - classification)[58], whether they are ulcerated and in case of excision whether this was completed or not.

Post-procedure

Adenoma detection rate (ADR) is currently the benchmark of quality in colonoscopy and represents the number of colonoscopies at which one or more histologically confirmed adenomas were found divided by the total number of colonoscopies performed in the same time period[59]. ADR reflects a colonoscopist’s technical skills and care to achieve visualization of the entire colon during the procedure. High ADRs reduce the probability of interval cancer by correctly identify surveillance intervals[60]. The overall prevalence of CRC, polyps and adenomas may differ between patient populations according to gender, race, diet or environmental factors and subsequently ADRs may vary[61]. Measurement of ADR is greatly assisted by a direct link between the databases of the endoscopy and pathology departments, but this is not available everywhere[62]. Polypectomy rates can potentially provide an ADR estimate based on previous ADRs but polyp detection rate (PDR) should be used cautiously for polyps of the left colon[63-66]. Previous reports argue that reliability of ADR is much higher when refers to a sufficient volume of colonoscopies (> 150/year in our BCSP) while the number and features (size, histology or grade of dysplasia) of adenomas detected per procedure is not included when counting ADR[67,68]. The mean number of adenomas per procedure (MAP) (defined as the total number of adenomas detected divided by the number of procedures) and the mean number of adenomas per positive procedure (MAP +) (defined as the total number of adenomas detected divided by the number of procedures in which one or more adenomas were detected) can provide additional information for endoscopist’s performance[44,69,70]. We recommend an ADR > 15% as the minimum outcome unadjusted for gender or race.

Τhe reliability of a colonoscopy service is dependent on a well-organized aftercare system. This should provide patients with easy-access to further care pathways deemed necessary by colonoscopy such as appropriate time for follow-up colonoscopy (indicated by current guidelines) need for radiological or surgical examination or referral to local Multi-Disciplinary-Team (MDT) meeting. This network should ensure that no patient is lost to follow-up and it requires good communication between relevant departments (Gastroenterology, Radiology, Histopathology and Surgery).

A routine policy of contacting patients within a defined period of time (30 d) following colonoscopy is recommended to check for delayed adverse events related to the procedure and to obtain the overall patient’s feedback for the service. A simple quality questionnaire for each part of colonoscopy service is useful to detect problems with the service. We recommend a routine 30-d check for every patient having a colonoscopy while patients should also be encouraged to report any AE in the meantime. Regular reviews of complications and 30-d mortality is an essential part of quality assurance. Records of adverse events should be kept active. Clusters of AEs should instigate a formal review of individual cases.

CONCLUSION

Quality in colonoscopy encompasses optimal collaboration of various professionals with clearly defined processes. Quality assurance in colononoscopy should be based on measurement of simple and reproducible outcomes which permit regular checks on each step of the colonoscopy service. CIR and ADR are the key elements of personal endoscopic performance and their value is maximized when standards of patient’s safety, comfort and satisfaction are adequately monitored and reviewed.