Identifying Risk Factors Leading to Suboptimal Bowel Preparation

Abstract

Introduction: Suboptimal (incomplete or poor) bowel preparation (SOBP) can severely lower adenoma detection rate (ADR) with reported miss rates as high as 45%. Our study aims to identify risk factors that could lead to SOBPs in a multiethnic urban patient population. Methods: We conducted a single-center retrospective chart review of 323 patients with SOBPs who underwent screening colonoscopy at our center over a period of 5 years. We compared this group with regards to their comorbidities to a control group of 395 randomly selected patients who had optimal (good to excellent) bowel preparation (OBP) in the same time period. All patients received standard preparation using polyethylene glycol solution and 4 tabs of bisacodyl. Results: Of our diabetic patients 65% (n=128) had SOBPs vs. 35% (n=70) having OBPs, giving an OR of 3.0 (95% CI 2.2, 4.3), indicating that diabetics are 3 times more likely to have SOBPs. In our hypertensive patients 52% (n=194) had SOBPs vs. 47% (n=174) having OBPs, giving an OR of 1.9 (95% CI 1.4, 2.6) indicating with hypertension are approximately 2 times more likely to have SOBPs. Of our morbidly obese patients 92% (n=46) had SOBPs vs. 8% (n=4) having OBPs, giving an OR 16.4 (95% CI 5.7, 45), indicating that those with morbid obesity are 16.4 times more likely to have SOBPs. Lastly in our liver cirrhosis patients 85% (n=6) had SOBPs vs. 14% (n=1) who had OBPs, giving an OR 7.5 (95% CI 0.9, 62.5), indicating that cirrhotics are 7.5 times more likely to have SOBPs. Conclusion: Our data reveals that diabetes, hypertension, morbid obesity, and liver cirrhosis are risk factors associated with SOBP. With the implementation of the Affordable Care Act and the focus on costeffective quality care, this data is significant as it allows us to risk stratify patients prior to scheduling initial colonoscopies. Such individualized preparation instructions based on comorbidities will ultimately result in higher ADR with lower associated costs.Table 1Table 2