Post Polypectomy Bleeding. Que Sera, Sera? Whatever Will be, Will be?

Why should one be concerned about delayed post polypectomy bleeding (DPPB)? One in 40 patients with polypectomy for large polyps will have it. A large systematic review of 50 studies including 6779 large polyps > 20mm resected by polypectomy, endoscopic mucosal resection (EMR) or endoscopic submucosal dissection (ESD), "en bloc" or in piece-meal have had an DPPB incidence of 2.6% (2.1% - 3.2%) [1].

Cand we generally predict what group of patients are more likely to have DPPB? Jaruvongvanich et al [2] have designed a meta-analysis of 12 studies on 14,313 patients. The pooled DPPB rate was 1.5% (0.7% - 3.4%). Cardiovascular disease (OR = 1.55), hypertension (OR = 1.53), polyp size > 10 mm (OR = 3.41), and polyps located in the right colon (OR = 1.60) were identified as significant DPPB risk factors, whereas age, sex, alcohol use, smoking, diabetes, cerebrovascular disease, pedunculated morphology, and carcinoma histology were not.

Huai et al [3] have designed a nomogram for DPPB prediction, based on 6 parameters, with a high AUC of 0.824 in the validation model, as in Figure 1, below. The model states that older men on anticoagulant, with pedunculated polyps on the right side are likely to bleed more often. Age, gender, and pedunculated morphology were included in the model, as risk factors, as opposed to previously mentioned paper [2]. The jury is probably still out for certain risk factors.

Yet, at the individual level, what could a patient do? Could a form of active surveillance for DPPB may be proposed?

Here's an idea. Fecal Immunochemical Test (FIT). FIT can detect minute quantities of hemoglobin in faces. It is used for colorectal cancer population screening in asymptomatic average risk individuals (eg. 50 to 75 years old) or in primary care in certain groups of symptomatic individuals, without overt rectal bleeding, as an adjunct decision tool for colonoscopy evaluation [5]. A recent meta-analysis on more than 6 million asymptomatic average risk individuals of which more than 2.6 million completed a screening FIT test have found a 5.4% positivity rate, an 85% colonoscopy completion rate, a 47.8% adenoma detection rate, a 25.3% advanced adenoma detection rate and a 5.1% colorectal cancer rate [6].

The fecal hemoglobin (f-Hb) cut-offs are set at 10 or 20 microg/g feces in majority of screening programs, depending on colorectal cancer incidence and the healthcare system's colonoscopy availability [7]. FIT result is usually read in central laboratory by specialized machines with cut-off threshold set by the health policy makers.

Mowat et al [8] have used FIT tests differently. They studied the dynamic of FIT values, before and after colonoscopy with polypectomy. They have found that the median pre-polypectomy f-Hb was significantly greater in advanced adenomas as compared to non-advanced adenomas (6.0 vs. 1.0 μg/g feces, p < 0.0001). For patients with both pre and post polypectomy FIT, the dynamic showed a significant decrease for advanced adenomas (19.2 microg/g pre-polypectomy decreasing to 3.5 microg/g post-polypectomy, p=0.01) and remained fairly constant and low for non-advanced adenomas (0.8 microg/g pre-polypectomy to 1.0 microg/g post-polypectomy, p = 0.96). To note that the authors recommended patients that the second FIT test be taken after 3-weeks to allow healing of colorectal resection site, an interval arbitrarily chosen based on experience. This would theoretically avoid false positive elevated quantitative FIT test values due to occult bleeding form the resection site.

Maybe we can use FIT for DPPB surveillance. There are, however, few obstacles:

• The time to epithelization of a resection site is unknown, possibly 3-weeks as Mowat et al [8] have assumed. It depends on lesion size and whether the resection site has been closed with metallic clips.
• The occult bleeding from the resection site depends on time to epithelization, the prophylactic coagulation of vessels at the end of the procedure (snare tip, APC, CoagGrasper), the use of hemostatic gel (Purastat).
• The probability of positive FIT depends on the occult bleeding from the resection site, location (rectum versus right colon) and FIT cut-off value.
• FIT laboratory central reading for quantitative tests takes a certain amount of time (post office time) which defies the purpose of a rapid result for an immediate individual predictive information.
• DPPB mostly occurs in the first week after polypectomy and extremely rarely in the second week after polypectomy [4].

There are a lot of known unknowns and unknown unknowns.

We propose to use a rapid qualitative FIT test to be self-administered daily by the patient in the first 2 weeks after colonoscopy, after each passage of feces.

This will hopefully identify a desirable dynamic pattern of FIT results:

• Positive daily until day X, negative afterwards
• Positive daily until day X, negative afterwards, but then again positive in day Y just before DPPB We could hypothetically, do a rapid test in the day Y and if negative maybe leave the patient go about his usual activities, hobbies, and holidays.

The proposed qualitative rapid FIT is HEMOTRUST (Biosyntex, Switzerland, www.biosyntex.com). The cut-off for HEMOTRUST is 6 microg/g, which is lower than advanced adenoma pre-polypectomy values and higher than post-polypectomy values for both advanced and non-advanced adenomas [8].

Aim of the research project We aim to study the DPPB predictive value of FIT after resection of advanced colorectal adenomas.

Question of the research project Is there a dynamic of f-Hb presence after resection of advanced adenomas? Is this dynamic different in patients with DPPB after resection of advanced adenomas? Could we predict DPPB by studying f-Hb dynamic using qualitative rapid FIT?

Hypothesis The f-Hb assessment dynamic using qualitative rapid FIT may help predict DPPB at the individual level.