Colonoscopy Practice and Quality: Colonoscopy Practice and Quality: Polyp Pathology

0:00Welcome to Board Pearls. This is episode three of five of the Colonoscopy Practice and Quality chapter, in the Endoscopic Procedures module. This episode is what the pathology report decides, and when to look again. The first patient is a man whose polyp contained cancer, and his case shows why intramucosal adenocarcinoma isn't treated as cancer in the colon, how the depth and level of invasion are described, which four histologic features send a patient to surgery, and why suspicious lesions come out in one piece. A frail woman with deep invasion shows why depth alone is no longer an automatic colectomy. The woman from episode two shows the twenty twenty surveillance intervals, what happens when findings point to different intervals, and why a piecemeal resection site has its own schedule. The woman with an interval cancer from episode one shows how her tumor's MLH1 result separates sporadic cancer from Lynch syndrome. And a man with many serrated lesions shows what serrated polyposis requires.

0:57Here's the first patient. He's the forty-seven-year-old man from episode one. His twenty-two-millimeter pedunculated sigmoid polyp was removed in one piece with a hot snare after a detachable loop was placed at the base. The pathology reads: tubulovillous adenoma with a focus of invasive adenocarcinoma invading into the stalk, Haggitt level two, well differentiated, no lymphovascular invasion, low-grade tumor budding, and two millimeters from the cauterized stalk margin.

1:27He has a malignant polyp, a T one colorectal cancer removed at colonoscopy, and whether he needs surgery starts with what invasive means in the colon. Submucosal invasion, not mucosal involvement, carries metastatic potential, because the colon's lymphatics begin at the muscularis mucosae, so a neoplastic cell above that layer can't enter them. That's why intramucosal adenocarcinoma, also called T one a or s m zero, meaning dysplastic glands in the lamina propria that haven't breached the muscularis mucosae, behaves clinically as high-grade dysplasia. Complete endoscopic resection cures it, and surgery would be unnecessary harm. Had the nodule from the woman in episode two read intramucosal adenocarcinoma with clear margins, she'd need surveillance, not colectomy, despite the reflex the word adenocarcinoma triggers. True colorectal cancer requires submucosal invasion, T one or beyond, where lymphatics begin, and the question then becomes the risk of nodal spread.

2:26His invasion is described by depth, using the system that fits his polyp's shape. The Kikuchi system applies to sessile lesions and divides the submucosa into thirds: s m one, the superficial third, less than one thousand microns from the muscularis mucosae; s m two, the middle third; and s m three, the deep third, beyond one thousand microns. The Haggitt system applies to pedunculated polyps: level one invades the head, level two the neck, level three the stalk, and level four the submucosa of the bowel wall below the stalk, which is roughly equivalent to deep submucosal invasion in risk. A stalk puts distance between the head and the bowel wall, which Haggitt measures by landmarks, while a sessile lesion has no stalk, so Kikuchi measures submucosal depth directly. Deep invasion, meaning beyond one thousand microns, s m three, or Haggitt level four, was traditionally an automatic indication for surgery, because nodal risk was assumed to be high. His level two isn't in the deep group.

3:29Four high-risk histologic features decide surgery, and any one is enough. Poor differentiation, grade three or higher, marks aggressive biology. Lymphovascular invasion is direct evidence that tumor has entered lymphatic or vascular channels. High-grade tumor budding means ten or more buds, where a bud is a single cell or a cluster of up to four cells at the invasive front, counted in one hotspot of zero point seven eight five square millimeters. And a positive margin means less than one millimeter of clear tissue between tumor and the cut edge. When any one of these is present, surgery is strongly indicated regardless of the depth of invasion; they don't need to add up. He has none of the four: his tumor is well differentiated, with no lymphovascular invasion, low-grade budding, and a two-millimeter margin. So his resection in one piece is curative, and he doesn't need a sigmoid colectomy. Lymphovascular invasion, a half-millimeter margin, poor differentiation, or twelve buds would each have sent him to surgery, even at level two.

4:31Deep invasion on its own no longer forces surgery, and the second patient shows when that matters. An eighty-four-year-old woman with severe heart failure, chronic kidney disease, and oxygen-dependent chronic obstructive pulmonary disease had a twenty-five-millimeter sessile rectosigmoid lesion removed in one piece. It's adenocarcinoma invading the deep third of the submucosa, Kikuchi s m three, beyond one thousand microns, but it's well differentiated, with no lymphovascular invasion, low-grade budding, and three-millimeter clear margins, and her surgeon estimates a high operative risk. A large cohort and later series showed that deep submucosal invasion alone, without other adverse features, carries only about a two and a half percent risk of lymph node metastasis. So deep invasion alone isn't sufficient reason for surgery in a poor surgical candidate, and the decision weighs the risk factors together rather than treating depth as a single cutoff. Her operative risk far exceeds her small nodal risk, so after a frank discussion she's managed with endoscopic surveillance.

5:35Both decisions depended on the polyp coming out in one piece. Depth, level, margin, budding, and lymphovascular invasion all need an intact specimen, and piecemeal removal makes accurate staging impossible. Had his polyp come out in four pieces, the pathologist could have found cancer but couldn't have said how deep it went or whether the margin was clear. So the decisive step in his care happened at the scope: a lesion that looks malignant is resected for the pathologist, not just for a clean defect. When surgery is needed, staging and treatment of colorectal cancer are covered in chapter fifteen.

6:10Once the pathology has decided whether surgery is needed, it also sets when the next colonoscopy happens, and the woman from episode two shows how. She's sixty-three, and her thirty-five-millimeter granular mixed nodular cecal lesion was removed with the nodule in one piece and the rest piecemeal, with the margins treated with snare-tip soft coagulation. Her pathology shows tubulovillous adenoma with high-grade dysplasia in the nodule, with clear margins, plus a six-millimeter tubular adenoma from her transverse colon.

6:41Her next colonoscopy is set by the twenty twenty United States Multi-Society Task Force rules, which calibrate the interval to the highest-risk lesion at the index colonoscopy, so intervals shorten as risk rises. A normal colonoscopy, or up to twenty hyperplastic polyps under ten millimeters confined to the rectosigmoid, returns at ten years, because the risk is at the population baseline. One or two tubular adenomas under ten millimeters with low-grade dysplasia get seven to ten years, and three or four get three to five years. Three years applies with any one high-risk feature: five to ten adenomas, an adenoma of ten millimeters or more, tubulovillous or villous histology, high-grade dysplasia, a sessile serrated lesion of ten millimeters or more, a sessile serrated lesion with cytologic dysplasia, or a traditional serrated adenoma. More than ten adenomas gets one year and should prompt evaluation for a hereditary syndrome.

7:41Serrated lesions follow a similar pattern. One or two sessile serrated lesions under ten millimeters get five to ten years, and three or four get three to five years. A hyperplastic polyp of ten millimeters or more gets three to five years, because any large serrated polyp raises neoplasia risk, and pathologists disagree on hyperplastic versus sessile serrated histology at that size.

8:02Her two findings point to different intervals, and the shorter one applies, because the highest-risk lesion governs what comes next. Her six-millimeter adenoma alone would mean seven to ten years. But her cecal lesion was thirty-five millimeters, tubulovillous, and high-grade, and each of those features alone means three years, so her colon-wide surveillance is at three years.

8:23Her piecemeal resection site needs its own schedule, because the risk there is different. Piecemeal endoscopic mucosal resection, EMR, of any lesion twenty millimeters or larger triggers surveillance of the site at six months, then a year later, then three years after that. Its purpose is to find residual or recurrent adenoma at the scar while it's still small enough to resect endoscopically, before it grows into an interval cancer. Local recurrence after piecemeal EMR of large lesions is about fifteen to twenty percent without adjunctive techniques, and snare-tip soft coagulation of the margin lowers it to five to ten percent, which isn't zero, so the six-month look still applies after a careful resection. So two schedules run for her: the site checks at six months, eighteen months, and three years after that, and the colon-wide colonoscopy at three years, both written into her plan. A small, scarred recurrence at six months would suit underwater EMR, the technique for scarred or fibrotic lesions covered in episode two.

9:28The rules are easiest to learn one finding at a time. Two four-millimeter tubular adenomas mean seven to ten years. Four small tubular adenomas mean three to five. Seven small adenomas mean three years. Twelve adenomas mean one year and a hereditary syndrome evaluation. A single eleven-millimeter sessile serrated lesion means three years. Two seven-millimeter sessile serrated lesions without dysplasia mean five to ten. A twelve-millimeter hyperplastic polyp in the ascending colon means three to five. Many of these turn on ten millimeters: below it, small lesions step through long intervals by count, while at or above it a single lesion can mean three years, so a nine-millimeter and an eleven-millimeter adenoma sit in different places, and accurate measurement matters. And after an inadequate preparation, as episode one covers, none of these intervals apply; the colonoscopy is repeated within a year.

10:27The pathology report can also raise a hereditary question, and the woman from episode one shows how it's answered. She's the sixty-six-year-old with a flat ascending colon adenocarcinoma found three years after a reportedly normal colonoscopy. Immunohistochemistry on her tumor shows loss of MLH1 and PMS2.

10:48Her MLH1 loss could mean Lynch syndrome or the sporadic serrated pathway, and the answer decides whether her family needs testing. The serrated pathway, marked by BRAF V600E, a high CpG island methylator phenotype, and MLH1 promoter methylation, produces fifteen to thirty percent of colorectal cancers. Loss of MLH1 with PMS2 is either Lynch syndrome or sporadic methylation. A BRAF V600E mutation, or MLH1 promoter hypermethylation when BRAF is wild type, means sporadic cancer, and only when both are negative does the patient go to germline testing. Loss of MSH2, MSH6, or isolated PMS2 goes straight to germline testing, because sporadic methylation doesn't explain it. Her tumor carries BRAF V600E and MLH1 methylation, so it's a sporadic serrated-pathway cancer, and her children don't need Lynch testing on the basis of her tumor. Universal tumor testing is covered in chapter fifteen, and Lynch syndrome in chapter thirty-six.

11:55The serrated pathway matters to the colonoscopist in one more way, because many serrated lesions can add up to a syndrome. A fifty-two-year-old man has seven sessile serrated lesions in his right colon, all six millimeters or larger, two of them twelve millimeters. Under the twenty nineteen World Health Organization criteria, serrated polyposis syndrome is five or more serrated lesions proximal to the rectum, all at least five millimeters, with two or more at least ten millimeters, or more than twenty serrated lesions of any size with at least five proximal to the rectum. He meets the first criterion. Serrated polyposis is managed by colonoscopy every one to two years with clearance of every lesion of five millimeters or more, and his first-degree relatives start screening earlier.

12:42Here are the traps the boards set in malignant polyps. Intramucosal adenocarcinoma in the colon isn't sent to surgery. The colon's lymphatics begin at the muscularis mucosae, which is why invasion depth matters. Kikuchi levels describe sessile lesions and Haggitt levels describe pedunculated ones. Haggitt levels two and three aren't treated like level four. Deep invasion alone isn't an automatic colectomy in a poor surgical candidate. Any single high-risk feature, poor differentiation, lymphovascular invasion, high-grade budding, or a margin under one millimeter, is enough for surgery. And a lesion that looks malignant isn't removed piecemeal.

13:23For surveillance, the traps are these. The interval is set by the highest-risk lesion, not the most common one. A single adenoma of ten millimeters or more means three years. Villous histology or high-grade dysplasia alone means three years. More than ten adenomas means a hereditary evaluation. Piecemeal EMR of twenty millimeters or more needs a six-month site check. A large hyperplastic polyp means three to five years, not ten. Standard intervals don't apply after an inadequate preparation. MLH1 loss with BRAF V600E is sporadic and doesn't need germline testing, while MSH2 loss goes straight to germline testing. And serrated polyposis needs colonoscopy every one to two years.

14:09So, start to finish. Intramucosal carcinoma in the colon can't metastasize, so complete endoscopic resection cures it. Describe T one depth by Kikuchi for sessile lesions and Haggitt for pedunculated ones. Send to surgery for any one of poor differentiation, lymphovascular invasion, high-grade budding of ten or more, or a margin under one millimeter. Deep invasion alone carries about a two and a half percent nodal risk and doesn't mandate colectomy in a poor surgical candidate. Resect suspicious lesions in one piece.

14:44Set the interval by the highest-risk lesion: ten years for a normal examination or small distal hyperplastic polyps, seven to ten for one or two small adenomas, three to five for three or four, three years for five to ten adenomas, an adenoma of ten millimeters or more, villous histology, or high-grade dysplasia, and one year plus genetic evaluation for more than ten. A sessile serrated lesion of ten millimeters or more, one with dysplasia, or a traditional serrated adenoma means three years, and a large hyperplastic polyp three to five. After piecemeal EMR of twenty millimeters or more, check the site at six months, a year later, and three years after that. With MLH1 loss, BRAF V600E or methylation means sporadic cancer, and MSH2, MSH6, or isolated PMS2 loss goes straight to germline testing. Serrated polyposis needs colonoscopy every one to two years.

15:43Episode four is what can go wrong after polypectomy: bleeding, perforation, post-polypectomy syndrome, and the spleen.

15:51For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode three of five of chapter thirty, and I'll see you in the next one.