Predictors of Poor Bowel Preparations and Colonoscopy Cancellations in Inpatient Colonoscopies, a Single Center Retrospective Study

Background: An adequate bowel preparation prior to colonoscopy is a major quality-limiting factor that determines both the diagnostic and therapeutic yield of a colonoscopy. Colonoscopy is considered the gold standard for colon cancer screening and it is the primary approach to the workup of hematochezia, diarrhea and iron deficiency anemia (IDA). Several modifiable factors of bowel prep adequacy have been identified, that account for around 25% of inadequate bowel preparations in outpatient colonoscopies. However, the literature is sparse when examining the factors associated with inadequate preparations and procedure cancellations in an inpatient hospital setting. We aim to identify factors that affect bowel preparation adequacy and procedure cancellations among diagnostic colonoscopies performed during hospitalization.

Methods: We retrospectively reviewed the electronic medical records of 1,500 consecutive patients who had a diagnostic colonoscopy as an inpatient at a tertiary level hospital over a 2-year period. All patients were administered a clear liquid diet the day prior to the colonoscopy. Patients were then instructed to drink 4 L of polyethylene glycol (PEG, Golytely) between 5 am to 9 am on the day of the procedure. The clinical course of each case was followed to identify quality of preparations, cancelled procedures and the reasons for cancellations. We applied univariate and multivariate logistic regression analysis to identify variables to predict cancellation and poor preparation.

Results: A total of 1,029 patients were included in the study. 194 (18.8%) patients had colonoscopy cancellations and 268 (26.0%) had poor bowel preparations. Multivariate analysis revealed these factors to be associated with colonoscopy cancellations: education at the graduate school level [odds ratio (OR) =1.93, P=0.04], Hispanic ethnicity (OR =0.47, P<0.01), hemoglobin level <10 g/dL (OR =1.41, P=0.05) and if the colonoscopy was done for other indications (OR =0.53, P=0.04). Factors associated with poor bowel preparation on multivariate analysis, were dementia (OR =2.44, P=0.02), gastroparesis (OR =3.97, P=0.01) and inpatient opioids use (OR =1.69, P=0.04).

Conclusions: The rate of colonoscopy cancellations and poor bowel preparations in inpatient colonoscopies were high, and we were able to identify predictors of inadequate colon preparation and procedure cancellations. Exploring more individualized colon preparation regimens based on personal risk factors could reduce the number of inadequate and cancelled colonoscopies in an inpatient setting.