Adults with intellectual and developmental disabilities (IDD) frequently have comorbidities that might interfere with colonoscopy preparation and examination. In this article, the authors review their experience with colonoscopies performed from 2002 through 2010 on adults with IDD at a state institution to evaluate quality and safety of examinations and to formulate a plan to improve quality. In addition to IDD, these individuals had associated comorbidities, including digestive motility disorders and epilepsy. Results were compared with examinations performed in a random group of 40 adults without IDD examined between 2004 and 2008. The authors discovered a high rate of colonoscopy preparation failure and failed examinations in the IDD group. Preparation complications were also noted. The authors recommend careful preoperative evaluation of individuals with IDD and limiting colonoscopy to individuals who are able to tolerate preparation and comply with hydration.
Colonoscopy has become the preferred method for evaluation of lower digestive symptoms and detection and treatment of colon pathology. In the United States, it is one of the most commonly performed invasive procedures, with an estimated 14.2 million examinations performed in 2002 (Seeff, Richards, Shapiro, & Nadel, 2004). Colonoscopy is now the standard in the evaluation of occult and overt gastrointestinal (GI) tract bleeding, changes in bowel habit, inflammatory bowel diseases, iron deficiency anemia, and abdominal pain syndromes. Colonoscopy is also considered to be the gold standard for colorectal cancer screening and surveillance. Colorectal cancer is the third most common cancer and the second most common cause of cancer-related deaths in the United States (“Screening for Colon Cancer,” 2008). Most colorectal cancers arise over time in preexisting benign colon polyps. The National Polyp Study demonstrated that the incidence of colorectal cancer could be substantially reduced by removal of benign colon polyps at the time of colonoscopy (Winawer, Zauber, Ho, O'Brien, & Gottlieb, 1993). Subsequent retrospective studies (Lieberman, Weiss, Bond, Ahnen, & Garewal, 2000; Schoenfeld, Cash, Flood, Dobhan, & Eastone, 2005) demonstrated that asymptomatic adults over age 50 have a 20%–38% risk of having one or more benign adenomas, up to an 8% risk of advanced adenomas, and up to a 1% risk of invasive cancer at the time of initial screening colonoscopy. These studies support the concept of colorectal cancer (CRC) screening in asymptomatic average-risk adults.
There are three current sets of guidelines for colorectal cancer screening (Rex, Johnson, Anderson, Schoenfeld, & Burke, 2009; “Screening for Colon Cancer,” 2008; Smith, Cokkinides, & Brawkey, 2008). The 2008 American Cancer Society, U.S. Multisociety Task Force on Colorectal Cancer, and American College of Radiology guidelines (McFarland et al., 2008) divide testing into cancer detection tests and cancer prevention tests (Smith, Cokkinides, & Brawkey, 2008). A range of testing options is recommended from these two groups (Smith et al., 2008). The U.S. Preventive Services Task Force (2008) guidelines recommend a similar “menu of options,” including fecal occult blood testing, sigmoidoscopy, or colonoscopy for screening (“Screening for Colon Cancer,” 2008). The American College of Gastroenterology (2008) guidelines use an evidence-based system to rate the quality of the individual tests. Colonoscopy is recommended as the preferred cancer prevention test for screening and surveillance (Rex et al., 2009).
Colonoscopy quality, the ability to detect and remove adenomas and detect and diagnose carcinoma and other lesions, is directly related to the quality of preparation of the colon (Froehlich, Wietlishbach, Gonvers, Burnand, & Vader, 2005; Harewood, Sharma, & deGarmo, 2003). The colon must be completely cleansed of debris in order to evaluate as much of the mucosal surface as possible. Recent data suggest that although the incidence of colorectal cancer is decreasing, and this is likely related to improved screening techniques, the influence of colonoscopy on prevention of right-sided colon cancer appears to be negligible. This may be due to inadequate preparation of the right side of the colon, with retained debris obscuring small, flat, or depressed lesions in the right colon (Baxter, Goldwasser, Paszat, Saskin, & Urbach, 2008; Brenner, Hoffmeister, Arndt, Stegmaier, & Altenhofen, 2010). As anyone who has had a colonoscopy will likely be aware, complete cleansing of the colon is challenging and difficult, even under ideal circumstances (Oz, 2011). Compliance with colon lavage is difficult, and side effects of all available agents may interfere with preparation completion. It has been reported that up to 27% of preparations are inadequate in the general population (Froehlich et al., 2005).
There are four broad categories of agents used for colon cleansing: stimulants, enemas, osmotic agents, and polyethylene glycol–based (PEG) solutions. Stimulant laxatives such as senna and bisacodyl work by increasing colon peristalsis. Stimulants are not effective when used alone but are often used as adjuncts to preparation with osmotic agents and PEG lavage agents (Wexner, Beck, & Baron, 2006). Similarly, enemas are ineffective when used alone but can be used as adjuncts to preparation when there is retained debris in the rectosigmoid colon (Wexner et al., 2006). Osmotic agents, such as magnesium citrate and sodium phosphate (NaP), are hypertonic agents that work by drawing plasma water into the bowel lumen (Barkun, Chiba, Enns, Marcon, & Natsheh, 2006; Cohen, 2010; Wexner et al., 2006). PEG-based solutions contain a high molecular weight, nonabsorbable polymer in a dilute electrolyte solution, which is osmotically balanced. This solution is designed to pass through the colon with little or no absorption or secretion of fluid or electrolytes, thus washing the colon of debris (Barkun et al., 2006; Cohen, 2010; Wexner et al., 2006).
The ideal method of bowel preparation using the above categories of agents would reliably and safely cleanse the entire colon with low discomfort and side effects and with high tolerability and compliance. Unfortunately, this ideal method is yet to be determined. Two broad methods of preparation have been in common use in the United States: NaP (osmotic) products and PEG-based (lavage) solutions. Multiple comparative studies have demonstrated that these two methods have roughly equal efficacy in cleansing the colon (Belsey, Epstein, & Heresback, 2007; Tan & Tjandra, 2006). In clinical studies, patient tolerance and compliance were greater with NaP agents compared with PEG-based solutions, but side effects and complications were also greater with NaP. The use of NaP has been associated with hypovolemia, abnormalities of serum electrolytes, acute phosphate nephropathy, and renal failure. Because of safety concerns, the FDA issued an alert in December 2008, recommending that over-the-counter NaP products not be used for colon cleansing (U.S. Department of Health and Human Services, U.S. Food and Drug Administration, 2010) and requiring prescription NaP tablets to carry a “black box” warning regarding the risks of acute phosphate nephropathy. The warning specifically recommends against the use of NaP in certain high-risk patient groups (Salix Pharmaceuticals, 2009). Since December 2008, PEG-based solutions have been by far the most utilized products for preparation. These preparations differ with respect to volume of lavage, electrolyte content, molecular weight of the polymer, requirement for an adjunctive laxative, and timing of administration (Barkun et al. 2006; Cohen, 2010; Wexner et al., 2006). Split dosing, with half of the solution ingested the evening prior to examination and half of the solution ingested 4–6 hr prior to examination, has been shown to increase efficacy of cleansing and tolerability and is the preferred method of cleansing at this time (Cohen, 2010; Cohen, Kastenberg, Mount, & Safdi, 2009).
Often overlooked in bowel preparation is the importance of electrolyte balance and adequate hydration of the patient (Barkun et al., 2006; Cohen & Tennyson, 2010; Lichtenstein, Cohen, & Uribarri, 2007). Osmotic agents such as magnesium citrate and NaP draw fluid and electrolytes into the gut and will be ineffective and more likely to cause adverse events if the patient is not well hydrated. A clear liquid diet consisting of at least 2 L of clear liquids per day is recommended in conjunction with all available prep agents except the high-volume 4-L PEG-based solutions, to ensure efficacy and safety (Cohen & Tennyson, 2010). This is in addition to liquids ingested with the product itself. Thus, with a NaP split-dose tablet prep, a minimum of 3–4 L of clear liquids must be ingested on the day prior to examination (Parra-Blanco, Quintero, & Jimenez, 2007; Rex et al., 2006). Similarly, with commonly used low-volume PEG-based solutions, a minimum of 3–3.5 L of clear liquids are required on the day prior to examination (Di Palma, Rodriquez, McGowan, & Cleveland, 2009; Ell, Fischback, Bronisch, Dertinger, & Layer, 2009). Tap water enemas and soapsuds enemas must be administered carefully, as they may cause excessive water absorption resulting in hyponatremia.
Adults with intellectual and developmental disabilities (IDD) constitute a unique, high-risk group when considering colonoscopy, because of the likelihood of multiple comorbidities. These comorbidities might be expected to interfere with preparation tolerance and compliance and therefore to adversely affect the quality and safety of examination. Individuals with IDD may lack the capacity to consent to examination. Comprehension of and cooperation with preparation and examination may be severely limited or completely absent, depending on the level of intellectual disability. Adults with IDD may not be able to self-administer the preparation or may resist preparation, and there may be reluctance of others to administer preparation. There may be behavioral factors, such as self-injurious behavior, rumination, and vomiting, that interfere with compliance. Adults with IDD frequently have digestive motility disorders (Sullivan, 2008). Oropharyngeal dysphagia is common, often associated with aspiration, requiring dietary alteration or enteral feeding. Gastroesophageal reflux disorder (GERD), esophageal dysmotility, and gastroparesis are common, frequently associated with dysphagia, further complicating preparation administration. Constipation is common and frequently severe, often requiring administration of multiple oral and rectal agents to control. Finally, there is a high incidence of seizure disorders, which may be exacerbated by preparation administration.
We reviewed our experience with colonoscopy in adults with IDD at a state residential institution in Virginia from 2002 through 2010, to evaluate quality, efficacy, and safety of examinations. If individuals had additional examinations performed prior to 2002, these examinations were also included. We then compared results in this group with a group of outpatients without IDD, informally matched for age and gender but otherwise chosen at random from a computer list of procedures performed between 2004 and 2008. Fifty-eight colonoscopies were performed on 47 individuals with IDD. There were 31 men and 16 women, with a mean age of 53 years. The reasons for conducting a colonoscopy were CRC screening in 20 of the individuals, family history of colon neoplasms in 2, personal history of colon polyps in 9, overt rectal bleeding in 10, positive fecal occult blood test in 5, abdominal pain in 3, evaluation of anemia in 2, and change in bowel habit in 8. The IDD diagnosis was chronic encephalopathy in 33 (70%), Down syndrome in 5 (11%), and other genetic or congenital disorders in 9 (19%). Thirty (64%) individuals had profound ID, 11 (23%) had severe ID, 5 (11%) had moderate ID, and 1 (2%) had mild ID. Oropharyngeal dysphagia was present in 35 individuals (74%); 36 individuals (77%) had chronic constipation, classified as mild, moderate, or severe on the basis of the use of oral and rectal agents; 23 individuals (49%) had GERD; 8 (17%) had gastrostomy tubes; and 5 (11%) were dependent on enteral tube feeding. Finally, 22 individuals (47%) had epilepsy and were under treatment with anticonvulsant medications.
The IDD colonoscopy group included five individuals with Down syndrome. Individuals with Down syndrome constitute a unique group, as they have a substantially lower risk of solid tumors, including colorectal cancer. The association of Down syndrome with a low risk of colon cancer appears to be strong and is based on retrospective observational studies (Hasle, 2001; Hasle, Clemmensen, & Mikkelsen, 2000; Hill et al., 2003; Yang, Rasmussen, & Friedman, 2002). The three current sets of screening guidelines from the American Cancer Society, U.S. Multisociety Force on Colorectal Cancer (McFarland et al., 2008), and the U.S. Preventive Services Task Force (2008), and the American College of Gastroenterology (Rex et al., 2009) provide detailed reviews of screening for average-risk and high-risk groups but do not mention and do not provide any guidance for low-risk groups, such as individuals with Down syndrome. Probably as a result of this, most practitioners actually performing colonoscopy are unaware of the low-risk status of individuals with Down syndrome. Because no screening guidelines exist for individuals with low colon cancer risk, it seems prudent to discuss this as part of the informed consent process prior to proceeding with screening of individuals with Down syndrome.
Because of multiple comorbidities, nonstandard custom preparations for colonoscopy were used in all individuals with IDD. These preparations consisted of a 3-day clear liquid diet and combinations of products, including oral sodium phosphate (prior to December 2008) or PEG-based solutions, stimulant laxatives such as magnesium citrate or bisacodyl tablets, and sodium phosphate, tap water, and soapsuds enemas. Preparation product use was recorded in the medical record, but hydration was not recorded. Preparation quality was rated using an unvalidated system similar to the rating system proposed by Aronchick (Aronchick, Lipshutz, Wright, Dufrayne, & Bergman, 2000). Preparation was rated excellent if the mucosal surface was not obscured by debris, good if up to 10% of the surface was obscured in areas but could be cleared, fair if 10%–25% of the surface was obscured in areas and could not be cleared, and poor if more than 25% of the surface was obscured by debris. Excellent and good preps were considered adequate, and fair and poor preps were considered inadequate (see Table 1).
The control group consisted of 40 outpatients examined between 2004 and 2008. There were 15 women and 25 men, ranging in age from 25 years to 77 years, with a mean age of 53 years. The reasons for conducting an examination were colorectal cancer screening in 11 individuals, history of polyps in 6, family history of colon neoplasms in 9, rectal bleeding in 6, abdominal pain in 5, anemia in 2, and change in bowel habit in 1. Preparation information was available for 32 patients: 8 (25%) used aqueous sodium phosphate, 21 (66%) used 2-L PEG/electrolyte solution with adjunctive stimulant laxative, and 3 (9%) used a 4-L PEG/electrolyte solution.
In the IDD group, preparation was rated excellent in 8 (13.5%), good in 22 (37.3%), fair in 12 (20.3%), and poor in 15 (25.3%). Preparation quality was not recorded in one case, and colonoscopy was canceled in one case. Thus, in the IDD group, preparation was adequate in 51% and inadequate in 46% of examinations. In contrast, in the control group, preparation was rated excellent in 27 (67.5%), good in 11 (27.5%), and poor in 2 (5%) of examinations. Thus, in the control group, preparation was considered adequate in 95% of examinations and inadequate in only 5% of examinations. This represents an extraordinarily high rate of preparation failure in the IDD group, compared with the control group of adults without IDD. In the IDD group, there was one missed cecal adenocarcinoma because of poor preparation, which was detected on subsequent examination. There was one advanced adenoma not removed because of poor preparation, which required surgical resection. There was one patient who developed hyponatremia and subsequent cerebral edema. This individual was prepped with aqueous sodium phosphate and tap water enemas. Although numbers are small, this is an extraordinarily high rate of failed examinations when compared with colonoscopy examinations in the general population. As already noted, there was at least one serious complication associated with preparation. Other interesting findings were noted. In the IDD group, eight individuals (13.6%) were found to have redundant, dilated colons, compared with none in the control group. Twelve examinations were performed in these individuals, and preparation was judged adequate in six (50%), inadequate in five (42%), and not recorded in one. Four individuals with IDD who had two examinations each had both adequate and inadequate preparation on different occasions, suggesting that there may have been problems with preparation administration. Finally, eight IDD individuals with gastrostomy tubes had 10 colonoscopies, and preparation was judged adequate in eight exams (80%), inadequate in one (10%), and not recorded in one examination.
We reviewed colonoscopy preparations and examinations in adults with IDD at a state residential institution and found a high rate of preparation failure, accompanied by a high rate of inadequate examinations. One serious complication of preparation was noted. We reviewed our data and could find no definite link between inadequate preparation and any particular comorbidity. Rather, preparation failure appeared to be related to the inadequate administration of osmotic or PEG-based agents, accompanied by inadequate hydration. Because of multiple comorbidities, many IDD individuals could not tolerate or would not comply with adequate preparation. Individuals with excellent preps were more adequately hydrated and able to consume lavage solutions or osmotic agents, in addition to stimulant laxatives and enemas. Individuals with gastrostomy tubes were able to receive hydration and PEG-based solutions via their tubes, and preparations were generally adequate in this group.
Because of multiple comorbidities, adults with IDD constitute a unique population requiring special consideration when contemplating an invasive procedure such as colonoscopy. On the basis of our findings, we recommend careful preoperative evaluation of adults with IDD referred for colonoscopy examination and adherence to standard preparation techniques, including adequate hydration. Indications for examination and comorbidities should be carefully assessed. For elective examinations, such as colorectal polyp and cancer screening in average-risk individuals, the individual must be able to consume at least 3 L of clear liquids per day in order to be considered for colonoscopy examination. If the individual meets this requirement, our current practice is to provide standard preparation with a low-volume PEG electrolyte solution. Adjunctive stimulant or osmotic laxatives can be added as needed. The preparation period can be extended to 2 or even 3 days, depending on comorbidities such as oropharyngeal dysphagia or constipation. This is a preparation strategy frequently used in difficult-to-prep patients, but it has not been adequately evaluated in clinical trials (Wexner et al., 2006). Examinations are performed in the afternoon, to allow for split-dose administration and any adjustments to the preparation. A basic metabolic panel, calcium, and phosphorus levels can be obtained at baseline and repeated as needed. The staff or caregivers administering the prep call the physician at 4 p.m. on each preparation day and again at 8 a.m. on the day of examination to report progress and make necessary adjustments. The individual's routine regimen of enemas is continued if needed and can be altered by the physician if needed.
If the individual referred for elective examination cannot meet the fluid ingestion requirement or is considered too high risk for examination because of comorbidities, then other screening modalities are considered. Current practice is to perform annual fecal immunochemical testing (FIT) and flexible sigmoidoscopy every 5 years, a colorectal cancer screening strategy recommended in the American Cancer Society guidelines (McFarland et al., 2008). Individuals are referred for colonoscopy if a FIT test is positive without adequate explanation or if adenomas are detected at sigmoidoscopy. The goal here is to reduce the number of low-quality, high-risk colonoscopy examinations and to reduce the “number needed to scope” to find advanced adenomas and cancer (Atkins, Cook, Cuzick, Edwards, & Northover, 2002; Nakajima, Saito, & Soma, 2003; Van Rossem, Van Rijin, Laheij, & Van Oijen, 2008). This strategy may not have the cancer prevention power of colonoscopy, but it is preferable to performing inadequate colonoscopy examinations.
In the case of a nonelective or urgent examination in a high-risk individual, where oral administration of at least 3–4 L of fluid per day is not possible or is not considered safe, the individual should be hospitalized for nasogastric administration of a PEG-based solution and intravenous support of hydration and electrolytes or for closely supervised administration of sodium phosphate tablets or magnesium citrate with intravenous support of hydration and electrolytes. Neither of these methods of preparation has been adequately evaluated in prospective clinical trials (DeLegge & Kaplan, 2005; Marschall & Bartels, 1998; Wexner et al., 2006).
In conclusion, our review demonstrates that high-quality colonoscopy examinations can and should be performed in adults with IDD. However, adults with IDD constitute a unique high-risk group when considering colonoscopy and colorectal cancer screening. As with other special patient groups, such as those hospitalized (Chorev, Chadad, Segal, Shemesh, & Mor, 2007), the elderly (Seinela, Pehkonen, Laasanen, & Ahvenainen, 2003), severely constipated patients (Cohen & Tennyson, 2010), patients with GI bleeding (Green et al., 2005), and pediatric patients (Wexner et al., 2006), adults with IDD cannot be treated in the same way as the general population if a successful outcome is to be achieved. We recommend more individual consideration when deciding whether and how to proceed with colonoscopy and colorectal cancer screening in this group. If adequate preparation is not possible because of comorbidities, other screening methods should be considered. If colonoscopy must be performed, special arrangements must be made to achieve quality and reduce risk.
Editor-in-Charge: Glenn T. Fujiura
Leonard S. Fischer (e-mail: firstname.lastname@example.org), Gastrointestinal Medicine Associates, 3620 Joseph Siewick Drive, Fairfax, VA 22033, USA; Andrew Becker, Maria Paraguya, and Cecilia Chukwu, Northern Virginia Training Center.