A PREDICTIVE SCORE OF INADEQUATE BOWEL PREPARATION BASED ON A SELF-ADMINISTRED QUESTIONNAIRE

Aims:

The aim of this study was thus to develop a predictive score of inadequate bowel preparation, based on responses to a short self-administered questionnaire on potential risk factors, completed by patients during consultation prior to colonoscopy.

Methods:

In this single center study, consecutive patients with colonoscopy indication have been prospectively enrolled from February to November 2015. The primary outcome was inadequate bowel preparation defined by Boston Bowel Preparation Scale (BBPS) score < 7 or a score £ 1 for any of the 3 segments.

Results:

A total of 599 patients were included. The mean BBPS score was 7.6 ± 1.7, 147 (24.5%) colonoscopies were classified as inadequately prepared. Seven predictors were selected into the final prediction model of inadequate bowel preparation: diabetes or obesity (BMI ³ 30 kg/m²), irregular physical activity, liver cirrhosis, use of antidepressants or neuroleptics, use of opiate medication, history of intra-abdominal and/or pelvic surgery and history of inadequate bowel preparation. The total risk score had an AUROC [95% CI] of 0.621 [0.558 – 0.689], rate of correctly classified patients of 68.3%, a 45.7% sensitivity, 75.8% specificity, 38.6% PPV, and 80.8% NPV.

Conclusions:

A predictive score named “Prepa-Co” has been developed, it allows to classify patients according to their risk of inadequate bowel preparation. In clinical practice, this score could help adapt the prescription of the preparation.