Key points

Introduction

Despite its straightforward purpose and unpretentious name, bowel cleansing for colonoscopy remains a challenge for endoscopists and patients alike. Its objective to remove all solid and liquid debris from the colon lumen in order to make it possible for the operator to perform a thorough examination of the mucosal surface seems simple enough. Nonetheless, roughly 1 in 4 colonoscopies performed in the United States and western Europe are considered to have an inadequate-quality preparation. Even a thin film of mucus or chyme coating the mucosa may obscure a clinically important lesion. The introduction of split-dose regimens has significantly enhanced the ability to achieve high-quality cleansing, especially within the proximal colon, and has improved the detection rate of flat polyps that may be readily obscured by mucosal staining or a small pool of fluid.

Coincident with efforts to design a more effective dosing schema have been attempts to develop effective purgatives that are more acceptable to patients. Several lower-volume preparations have recently been approved in the United States. Notwithstanding these efforts, conventional 4-L polyethylene glycol (PEG) with electrolytes remains the benchmark by which the efficacy of other preparations should be compared.

In an effort to improve the overall quality of bowel preparation, and consequently the quality and effectiveness of colonoscopy, the US Multi-Society Task Force (MSTF) on colorectal cancer (CRC) now recommends that clinicians meet or exceed a target rate of 85% for adequate bowel preparation. While acknowledging that rates of adequate preparation are related, in part, to a variety of patient factors such as age, gender, comorbid disease, and socioeconomic status, and that differences in these and other variables from practice to practice can significantly affect the proportion of patients with an adequate preparation, the gastrointestinal professional societies think that it is possible for most practitioners in the United States to achieve this goal. However, for some endoscopists it will require a reformulation of their approach to bowel preparation with greater emphasis on patient education, increased patient activation, adoption of preparation agents that have track records for high rates of success, and the implementation of split-dose regimens. This article reviews recent updates to bowel preparation guidelines and the evidence supporting these recommendations.

Introduction

Despite its straightforward purpose and unpretentious name, bowel cleansing for colonoscopy remains a challenge for endoscopists and patients alike. Its objective to remove all solid and liquid debris from the colon lumen in order to make it possible for the operator to perform a thorough examination of the mucosal surface seems simple enough. Nonetheless, roughly 1 in 4 colonoscopies performed in the United States and western Europe are considered to have an inadequate-quality preparation. Even a thin film of mucus or chyme coating the mucosa may obscure a clinically important lesion. The introduction of split-dose regimens has significantly enhanced the ability to achieve high-quality cleansing, especially within the proximal colon, and has improved the detection rate of flat polyps that may be readily obscured by mucosal staining or a small pool of fluid.

Coincident with efforts to design a more effective dosing schema have been attempts to develop effective purgatives that are more acceptable to patients. Several lower-volume preparations have recently been approved in the United States. Notwithstanding these efforts, conventional 4-L polyethylene glycol (PEG) with electrolytes remains the benchmark by which the efficacy of other preparations should be compared.

In an effort to improve the overall quality of bowel preparation, and consequently the quality and effectiveness of colonoscopy, the US Multi-Society Task Force (MSTF) on colorectal cancer (CRC) now recommends that clinicians meet or exceed a target rate of 85% for adequate bowel preparation. While acknowledging that rates of adequate preparation are related, in part, to a variety of patient factors such as age, gender, comorbid disease, and socioeconomic status, and that differences in these and other variables from practice to practice can significantly affect the proportion of patients with an adequate preparation, the gastrointestinal professional societies think that it is possible for most practitioners in the United States to achieve this goal. However, for some endoscopists it will require a reformulation of their approach to bowel preparation with greater emphasis on patient education, increased patient activation, adoption of preparation agents that have track records for high rates of success, and the implementation of split-dose regimens. This article reviews recent updates to bowel preparation guidelines and the evidence supporting these recommendations.

Bowel preparation and its impact on colonoscopy quality

The quality of bowel cleansing at colonoscopy should be assessed during scope withdrawal, after all efforts at washing and suctioning fluid and debris from the lumen have been completed. Such an approach to grading largely discounts the presence of retained fluid and small bits of stool, because these can usually be removed during the procedure. Given its importance in improving the quality of colonoscopy, the time required for adequate washing should not be minimized. A recent prospective study found that the mean time spent washing/suctioning during colonoscopy was 4.1 minutes. As anticipated, the time and the volume of instilled fluid were inversely proportional to the baseline preparation grade, ranging from 2.7 to 9.2 minutes and 167 to 807 mL for excellent to poor preparations, respectively. In 33 (6%) patients, the quality of preparation was substantially improved, from fair/poor to good/excellent (or inadequate to adequate). This finding shows that intraprocedural washing and cleaning are integral components of the examination and can have an appreciable impact on the quality and cost-effectiveness of colonoscopy. The time allotted for each procedure should therefore be sufficient for the endoscopist to thoroughly cleanse those segments of the bowel that are not adequately prepared during the preprocedure period.

The quality of the bowel preparation should be recorded in a written report generated after each colonoscopy. Many clinicians use a nonstandardized 4-point scale (excellent, good, fair, and poor) to describe the quality of bowel cleansing, often without a precise definition of each grade. As a result, a preparation rated good by one endoscopist may be considered fair by another. Further, the adequacy of a fair preparation, in terms of adenoma detection, has yet to be fully defined. Nonetheless, many endoscopists shorten the interval between examinations when the quality of the bowel preparation is considered inadequate. For example, a retrospective review of more than 16,000 colonoscopies performed at an academic medical center found that nearly 60% of average-risk patients with normal colonoscopy results underwent early repeat examination because of a preparation quality rated as fair. A recent systematic review and meta-analysis of 11 studies observed no difference in adenoma detection rate (ADR) between examinations with intermediate-quality (fair) compared with those with high-quality preparation (good or excellent) (odds ratio [OR], 0.94; 95% confidence interval [CI], 0.80, 1.10). In contrast, ADRs were significantly higher with both intermediate-quality (OR, 1.39; 95% CI, 1.08, 1.79) and high-quality preparations (OR, 1.41; 95% CI, 1.21, 1.64) compared with low-quality (poor) preparations. Similar findings were observed in an analysis of 13,000 colonoscopies within a statewide registry, with comparable ADRs in examinations with fair-quality versus optimal-quality preparation. If confirmed by other investigators, these findings indicate that a fair-quality preparation could be sufficient for the detection of adenomas during screening and surveillance colonoscopy.

The current emphasis is on classifying a bowel preparation as either adequate or inadequate. An adequate preparation is one that allows the detection of lesions greater than 5 mm in size. As such, the term adequate implies that bowel cleansing is sufficiently good for the examiner to identify clinically important lesions. The US MSTF guidelines define an adequate preparation as one that enables the endoscopist to follow the recommended screening and surveillance guidelines.

Implications of an inadequate bowel preparation

Adenoma detection is currently the most important measure of colonoscopy quality ( Box 1 ). Data from a national endoscopic database indicated that an adequate-quality preparation resulted in a higher detection rate for adenomas of all sizes (OR, 1.21; 95% CI, 1.16, 1.25) compared with an inadequate preparation. Similar findings were observed in a prospective, multicenter European trial in which all polyps (OR, 1.46; 95% CI, 1.11, 1.93) as well as larger (>10 mm) polyps (OR, 1.72; 95% CI, 1.11, 2.67) were detected more often with a high-quality compared with a low-quality preparation. Flat lesions within the proximal colon are also more likely to be detected when the preparation quality is rated as good. For example, a single-center study reported significantly more flat polyps detected with a higher-quality (same-day) versus a lower-quality (day-before) preparation (21.6% vs 9%; P = .02).

In contrast, poor-quality bowel preparation leads to lower detection rates, which means more missed lesions. Two single-center studies from academic medical centers examined the rate of missed lesions with an inadequate bowel preparation on the index examination, based on the results of a follow-up procedure performed within 3 years of the baseline study. The overall adenoma miss rates were, respectively, 42% and 47.9%. Although the absolute numbers may differ from one center to another, these rates show that significant numbers of polyps, both large and small, are missed with an inadequate preparation.

Operational efficiency is also affected by the quality of bowel preparation. European investigators observed that cecal intubation time (11.9 vs 16.1 minutes; P <.001), instrument withdrawal time (9.8 vs 11.3 minutes; P <.001), and cecal intubation rates (90.4% vs 71.1%; OR, 3.33; 95% CI, 2.36, 4.70) were all better with high-quality than low-quality preparations. As indicated earlier, the intraprocedural time washing the bowel was prolonged with poor-quality preparations.

Endoscopists have long struggled to find the most appropriate follow-up strategy when the quality of bowel preparation is suboptimal. There is evidence of surveillance overuse in low-risk patients and well as underuse in high-risk patients. For example, a review of practice patterns among 152 physicians from 55 practices indicated that more aggressive surveillance strategies were recommended when the preparation quality was good/fair/poor compared with those rated excellent. In an effort to provide better guidance to clinicians, the 2012 MSTF guidelines state that, “If the bowel preparation is poor, the MSTF recommends that in most cases the examination should be repeated within 1 year. … If the bowel preparation is fair but adequate (to detect lesions >5 mm) and if small (<10 mm) tubular adenomas are detected, follow-up at 5 years should be considered.” In order to improve compliance with clinical practice guidelines, both governmental and commercial payers are developing quality measures for colonoscopy that reflect clinicians’ adherence with screening and surveillance guidelines. For example, the Physician Quality Reporting System, developed by Medicare, will begin using payment adjustments in 2015 to encourage physicians to report on specific measures related to follow-up intervals for screening and surveillance examinations. The message is clear: consistent high-quality bowel cleaning is a necessary element of effective and efficient colonoscopy. A rate of 85% has been proposed by the MSTF on CRC as a benchmark for adequate preparation on a per-physician basis.

Difficult-to-prepare patients

Several patient-related characteristics, including age, gender, body mass index (BMI), and hospital inpatient status, have been associated with the quality of preparation ( Box 2 ). The association with higher age and male gender has a low predictability level and is not consistently observed across all studies. In contrast, a large study of 1588 patients found that BMI greater than 25 was an independent predictor of inadequate preparation, with each unit increase in BMI more than 25 increasing the likelihood of inadequate preparation by 2.1%. Further, additional independent predictors of inadequate preparation, combined with high BMI, exponentially increased the risk of a suboptimal preparation.

Various medical conditions and medications are also associated with an increased likelihood of inadequate preparation. The list of conditions includes stroke and Parkinson disease, impaired mobility, diabetes mellitus, spinal cord injury, and prior gastrointestinal resection. The medications that have been implicated include narcotics and tricyclic antidepressants. Although the OR for any of these conditions alone adversely affecting the preparation is low, ranging from 1.1 to 3.2, a cumulative effect has been observed so that elderly patients with multiple comorbid diseases have a significantly increased likelihood of having an inadequate preparation. Several investigators have attempted to develop a predictive model using a composite score to increase the success in identifying individuals at risk. One such model had only modest ability to predict an inadequate preparation (area under the curve, 0.63), and its utility in clinical practice seems limited at present.

Additional predictors of inadequate preparation include health literacy and patient activation. Although other variables, such as health insurance status, socioeconomic status, and educational level, have also been observed to predict suboptimal bowel preparation in earlier studies, it is likely that the common denominator is reduced health literacy, which can lead to difficulty or inability to read the educational materials and instructions provided before colonoscopy, failure to comprehend the importance of bowel preparation in the overall success of colonoscopy, or lack of understanding regarding preventive health measures. Patient activation refers to the motivation or mind-set of a patient to participate, or be engaged, in their health care. Although seemingly interrelated, health literacy and patient activation are independent variables as predictors of inadequate preparation. In a recent study of nearly 500 outpatients followed either at an academic medical center clinic or a federally qualified health center, low levels of patient activation (OR, 2.12; 1.3–3.45), but not health literacy (OR, 0.76; 0.38–1.52), was an independent predictor of inadequate preparation. The solution to this problem, put succinctly in a recent editorial, is “education, motivation, and reminders.” This can be in the form of 1:1 teaching sessions, bowel preparation classes, instructional videos, online tutorials, or in some cases patient navigators. More research is needed to develop better methods for improving patient understanding and willingness to become actively involved in the preparation process.

Selecting a bowel preparation agent