In patients with iron deficiency anaemia (IDA), the diagnostic yield of gastroscopy and colonoscopy (bidirectional endoscopy) in detecting neoplastic lesions is low. This study aimed to develop and validate a faecal immunochemical test (FIT)-based model to optimise the work-up of patients with IDA.
Outpatients with IDA were enrolled in a prospective, multicentre study from April 2016 to October 2019. One FIT was performed before bidirectional endoscopy. Significant gastrointestinal lesions were recorded and a combined model developed with variables that were independently associated with significant colorectal lesions in the multivariate analysis. The model cut-off was selected to provide a sensitivity of at least 95% for colorectal cancer (CRC) detection, and its performance was compared to different FIT cut-offs. The data set was randomly split into two groups (developed and validation cohorts). An online calculator was developed for clinical application.
The development and validation cohorts included 373 and 160 patients, respectively. The developed model included FIT value, age, and sex. In the development and validation cohorts, a model cut-off of 0.1375 provided a negative predictive value of 98.1 and 96.7% for CRC and 90.7 and 88.3% for significant colorectal lesions, respectively. This combined model reduced the rate of missed significant colorectal lesions compared to FIT alone and could have avoided more than one-fourth of colonoscopies.
The FIT-based combined model developed in this study may serve as a useful diagnostic tool to triage IDA patients for early endoscopic referral, resulting in considerable reduction of unnecessary colonoscopies.