Malignant polyps – when to resect?

Paul S Rooney, Royal Liverpool

The management of malignant polyps is perhaps the most controversial areas in all colorectal surgery
A malignant polyp (MP) is one in which there has been invasion of the submucosa.
In the past units had policies to either treat all such polyps by radical surgery or treat all conservatively. Only in the last 15 years have there been enough data to let surgeons (physicians too!) make rational decisions.
Surgery for MP arises from one of four clinical scenarios;

1: The polyp cannot be treated endoscopically.

2: the polyp looks benign when treated but has malignant features on histology.

3:MP treated endoscopically but incompletely.

4:MP treated successfully but bad prognostic histological features.

It is the assessment of the risk of lymphatic nodal involvement (LN+ve) that vexes the surgeon.
LN+ve polyps may be as common as 14% where the muscularis mucosa has been breached. The rate of LN+ve depends on the type of polyp pedunculated TYPE I or flat TYPE II. Each has been assessed in longitudinal studies and classified to make things easy for us:

Haggitt for type I ; haggitt level 1-4 (1 good, 4 bad), level 4 need surgery (12% LN+ve).

Kukuchi for typeII (flat cancers);sm 1-3 (1good, 3 bad), sm3 need surgery (36%LN+ve).

Other factors to consider are tumour budding , grade ,and resection margin a studied by Ueno, who found the risk of LN+ve was 0.7%, 20%, 36% for no 1 or 2 risk factors respectively.

However one must also consider the morbidity and mortality (2-7.5%) of surgery balanced against the risk of metastatic disease.

What recommendations can we give?

Perhaps these: surgery for MP is not required when

1. The cancer is not poorly differentiated.
2. there is no vascular or lymphatic invasion
3. the diathermy margin of the MP is not involved with cancer
4. Haggitt 1-3
5. Kukuchi sm1 or sm2
6. there is no tumour budding
7. when the mortality of the procedure out weighs the risk of LN+ve

In our institution there has been an ad hoc arrangement based on the above criteria. This lead us to look examine all our surgical resection specimens with cancers less than 2cm in size. Of 1763 specimens only 61 (3.4%) were <2cm, 33% type1 and 64% type 2 (Flat). The rate of LN+ve was high in both groups 30% and 39%, reflecting the selective nature of the study and usefulness of the above guidelines.

References:

1. Hagitt RC et al Prognostic factors in colorectal carcinomas arising in adenomas Gastroenterology 1985:89(2):328-36
2. Kikuchi R et al Management of early invasive colorectal cancer risk of recurrence and clinical guidelines
3. Ueno H Risk factors for an adverse outcome in early invasive colorectal carcinoma. Gastroenterology 2004;127(2):385-94

Tweedle EM et al Small flat colorectal cancers in the UK population: an analysis of resected specimens Colorectal Disease 2007;9: 641-46