A new risk assessment tool based on pathology tests and demographic measures can markedly improve triage accuracy for colorectal cancer and help relieve overburdened colonoscopy clinics, Queensland clinicians say.
The triage tool developed by Dr Anton Lord and colleagues at the QIMR Berghofer Medical Research Institute in Brisbane is far superior to the current symptom-based triaging system for prioritising patients referred from primary care for colonoscopy, their evaluation in 428 patients has shown.
Building on the usefulness of faecal haemoglobin as a rule-out test for significant bowel disease, Dr Lord and colleagues added objective pathology markers and demographic measures to create a more accurate ‘rule-in’ assessment tool that improved sensitivity from 11.6% to 24.5% and which had a specificity of 96.5% .
“Notably the positive predictive value of those identified at high risk by the risk assessment tool was 74% compared to 30% for the current best practice,” they write in BMC Cancer.
The authors note that their hospital’s current median wait times from GP referral until colonoscopy were 69 days for category 1 (urgent) patients, 264 days for category 2 (semi-urgent) patients and 220 days for category 3 (non-urgent) patients.
And since symptoms were a poor indicator of significant bowel disease, this meant endoscopy clinics were doing a large number of unnecessary procedures on patients with no significant bowel disease, while patients with significant disease were having to wait for extended periods during which time their disease could worsen, they noted.
“These findings highlight the workload pressures on a major referral centre and demonstrate a service where demand is outstripping resources. The key issue is prioritisation of patients within categories 2 and 3 such that SBD diagnosis is not delayed,” they said.
In their evaluation, 1% of patients were found to have colorectal cancer, 26% had significant bowel disease (high risk adenoma or IBD) and 42% had a normal colonoscopy.
Since their risk assessment tool had a much higher positive prediction value for significant bowel disease, it could be used by gastroenterology department staff at triage clinics to more accurately identify patients to be prioritised for urgent colonoscopy, they concluded.
“Due to the objective nature of the risk assessment tool it could conceivably be used by support staff within a hospital setting reducing the burden on specialists,” they said.
The risk assessment algorithm includes a faecal immunochemical test (FIT) and a blood risk score based on triglycerides, glucose, magnesium and creatinine abnormalities. The demographic measures include age over or under 65.
“It is critical that individuals with high risk of having significant bowel disease are triaged to the appropriate category with the shortest wait time. Here we provide a framework where evidence-based risk factors can be included in the diagnostic pathway for significant bowel pathology. To the best of our knowledge, this is the first study identifying a combination of blood markers along with FIT and demographic markers have a higher diagnostic accuracy for SBD than FIT alone,” the authors concluded.