Predicting a Poor Prep: Is the Drink at Fault?

Study results of 90K+ patients presented at DDW 2017

Inadequate bowel preparation (IBP) reduces our ability to detect polyps, prolongs procedure time and raises electrocautery risk. Research to date suggests IBP occurs in about 1 in 4 patients, but we have relatively little information on the effect of specific bowel preparations on IBP in patients not in clinical trials.

Advertising Policy

Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy

Many FDA approved bowel preparations have been studied in healthy people. “The preps work in those people. But in the real world, we’re seeing a lot of variation in the quality of the prep in patients with comorbid illness,” says Cleveland Clinic Chief  Gastroenterology Fellow Ari Garber, MD, EdD.

And so, Dr. Garber, along with Maged Rizk, MD, Carol Burke, MD, and their Cleveland Clinic research team decided to use the institution’s electronic medical record database of more than 115,000 patients who had undergone colonoscopies between 2011 and 2016 to investigate the real-world quality of different bowel preps and possible predictors of IBP.

Study design

After excluding inpatients, those younger than 18 and those with missing bowel prep data, the study population was still quite large: 91,045 individuals. Of those, 41 percent had undergone colonoscopy for routine screening, while the rest were either for surveillance (18.2%) or diagnostic (41.2%) purposes.

The most commonly prescribed bowel preparation product was 4 liter PEG (77%), followed by 2 liter PEG with ascorbic acid (20.2%).

Overall, 16 percent of the patients had IBP, defined as a score of fair/poor as opposed to good/excellent on the Aronchik scale.

Advertising Policy

A breakdown of the findings

The IBP risk rose significantly with age, male gender, non-native English speaker, lower median household income and government health insurance (Medicaid and Medicare).

Patients with just about any major comorbidity were also less likely to achieve adequate prep, with nearly double the odds for IBP in those with cirrhosis, congestive heart failure, diabetes, dementia, stroke and myocardial infarction.

And compared to those who had the colonoscopy done for routine screening, the IBP odds were significantly greater for those having surveillance and diagnostic colonoscopies (by 10% percent and 30%, respectively).

A somewhat puzzling discovery

While none of the above findings are particularly surprising, the findings by prep product were a bit counter-intuitive: As a whole high-volume (4 liter) PEG preps were associated with a 14 percent higher IBP rate compared to the low volume (2 liter or less) preps. The consequences of IBP were significant, with lower detection of polyps and adenomas compared to those with adequate prep (40.2% vs. 44.6% and 25.6% vs. 26.7%, respectively).

The reason for IBP in outpatients prescribed 4 L preps isn’t clear, and it may be due to improved tolerability of low versus high volume prep. Low volume prep use could be a surrogate for different demographic, comorbid and socioeconomic features and an adjustment for cofounders is our next step. From what we have observed, “Less may be more in the outpatient setting,” Dr. Garber notes. However, these findings need to be further analyzed before we conclude the various contributions from the prep versus the patient.

Advertising Policy

Should prep be changed? More study needed

The important question now, he says, is “If these findings hold after further analysis and the lower volume preps are doing a better job and are better tolerated, why are we torturing patients with large-volume preps?”

This study is the first in a large ambulatory setting to examine the adequacy of colon prep based on the drinks patients were prescribed. Cleveland Clinic, a tertiary health care system with a large volume of patients seen per year, provided an ideal dataset to shed light on the issue, Dr. Garber says. “Only by being able to get data of this size are we able to answer questions of this magnitude.”