Annoying Pathology Rules

[pathstudent]

---

Members don't see this ad.

Was anyone else instructed to put in one section per cm for leiomyomas as the standard of grossing. I quickly came to think of it as the dumbest thing I have ever seen after seeing hundreds of benign leiomyomas. It is especially annoying when a woman has 15 lesions; how absurd to sample each one. In fact has there ever even been a 2cm leiomyosarcoma in a uterus? FOr christ's sake you can almost gross only a leiomyoma in a uterus.

What other rules are you asked to follow that seem baseless?

 

[KluverB]

one section per cm of lipoma.

 

[gbwillner]

whatever the deal is with TURP chip sampling. I already forgot the rule, but it sucks (something like entirely submit the first 10g, then 1 block per additional 5g).... really... why?

And 12 lymph nodes on an irradiated sigmoid/rectum. Yer kidding, right? Don't you know you are supposed to average 12 LNs per large bowel case? that means some will have less than that, so don't send me back to the bucket. Why not just submit all the fat?

And placentas. All of them. THey should be gross only.

 

[Perry Mason]

Well the average per large bowel case is actually more like 30 - 40. I agree it's a bummer to look for them post CRT, especially at the end of a busy day. Nevertheless, I had two cases in my first year only where one single node out of 30 or 40 turned out to be positive. And it wasn't a big firm grossly positive one either, just a little sucker you would never have thought much of.

 

[pathstudent]

one section per cm of lipoma.

roger that. Especially if it is subcutaneous from the arm or something like that. What a waste of eye-power.

 

[deleted314957]

30-40 in the average large bowel case? Far from my near 30 year experience.

 

[KCShaw]

If you know what it is, don't submit sections at all. The trouble is, most pathologists don't want to be the one who assumed something was as benign as it "always" is, only to find out the hard way that they were wrong. I.e., the hernia sac harboring a tiny mesothelioma, the gallbladder for chronic cholelithiasis with a small carcinoma, the placenta with metastatic melanoma (okay, this one was at least suspected), etc. Although I don't recall finding anything from sampling a panniculectomy.

So unless there is both strong evidence that it's -not- worth doing something, and surgeons aren't requesting it, we tend to go overboard (sampling, stains, etc.). Especially with remotely related tumors, when there is evidence of utility in one situation it tends to get broadly applied to all until proven otherwise. We're actually probably getting better about this as evidence based decision-making + cost saving demands are both more prevalent. But, basically, the pathologist's job is to not miss anything, never be wrong, and provide all possible useful information for treatment & prognostic purposes. Everyone else gets to use a "working diagnosis," "treat empirically," etc.

 

[malchik]

The colon lymph node count rule is easily the most annoying rule in surg path. I'm going to start sending my colons to perry mason so he can get 40 lymph nodes out of my 15 cm irradiated recta.

 

[2121115]

The colon lymph node count rule is easily the most annoying rule in surg path.

There is little data to back it up either.

 

[Perry Mason]

http://www.biomedcentral.com/1471-2407/10/267

I am not claiming that searching for lymph nodes in the colonic fat is enjoyable. My point is, whether you find 12 most obvious nodes and stop at that or continue squeezing fat can (and does) make a difference to the patient, unlike submitting leiomyomata and all those other examples above. Which kinda makes this (still annoying) rule not that stupid after all.

 

[KCShaw]

My impression has always been the same -- that node searches on primary resection for colon cancer was one of the reasonably well evidenced goals of grossing. It was one of the few things I ever got upset with residents about as chief, on the rare occasion it was clear they just didn't bother looking, or kept shouting "only X more until I can stop looking!" when they hadn't even gotten to tissue in the region of the tumor. (I guess they would rather have put in some GI biopsies or grossed a few fresh placentas? Early hospital food lunch?? I dunno. It's tedious, which could be said about most grossing, but at least there's a little challenge in the search.)

Now, when you start talking about post-treatment specimens, that may be another story. Again, the danger of applying relevance in limited circumstances to all circumstances may be at issue.

 

[pathstudent]

http://www.biomedcentral.com/1471-2407/10/267

I am not claiming that searching for lymph nodes in the colonic fat is enjoyable. My point is, whether you find 12 most obvious nodes and stop at that or continue squeezing fat can (and does) make a difference to the patient, unlike submitting leiomyomata and all those other examples above. Which kinda makes this (still annoying) rule not that stupid after all.

We were taught that you were required to find and submit all the lymph nodes. 12 were only required for adequate staging.

 

[zao275]

We were taught that you were required to find and submit all the lymph nodes. 12 were only required for adequate staging.

Agree. Same here.

However, submitting 100+ lymph nodes on a colectomy for ulcerative colitis (with no known carcinoma) is an annoying rule for me (not everyone does this). The reasoning that I have heard is that one case had a single node with tiny focus of cancer out of over 100 nodes from a colon with UC. There was no obvious primary in the colon. That seems like one of the rare circumstances that KCShaw mentions.

 

[2121115]

I'm not saying that we shouldn't get nodes on colon cancer specimens, just that the minimum number is arbitrary and sometimes (particularly with treated cases) they just aren't there.

 

[KCShaw]

Ah, there I'm with you. As I mentioned above, one of my pet peeves was residents announcing they were "done" when they had reached some given number, despite there being a significant amount of relevant yet completely unsearched tissue before them. It's one thing to miss a few nodes here and there -- quite another to decide to stop looking based solely on a number taken out of context. Most of these studies involved cases where "all" lymph nodes were submitted, or at least all relevant tissue searched, and statistical analyses then performed.

 

[yaah]

The colon lymph node thing is the classic example of drawing the wrong conclusions. Someone has published a study which states that finding more lymph nodes in colon cancer cases improves prognosis (or something like that). The lesson that should be learned is not that you HAVE HAVE HAVE to find 12 nodes! That doesn't make any sense. How much time the resident spends grossing in a case has absolutely no bearing on the patient's ultimate prognosis!

 

[Arctic Char]

whatever the deal is with TURP chip sampling. I already forgot the rule, but it sucks (something like entirely submit the first 10g, then 1 block per additional 5g).... really... why?

And 12 lymph nodes on an irradiated sigmoid/rectum. Yer kidding, right? Don't you know you are supposed to average 12 LNs per large bowel case? that means some will have less than that, so don't send me back to the bucket. Why not just submit all the fat?

And placentas. All of them. THey should be gross only.

we do 5 cassettes for the first 10 g, then one cassette for every 5 g thereafter. i've seen cancer lurking in there more than few times

 

[pathstudent]

The colon lymph node thing is the classic example of drawing the wrong conclusions. Someone has published a study which states that finding more lymph nodes in colon cancer cases improves prognosis (or something like that). The lesson that should be learned is not that you HAVE HAVE HAVE to find 12 nodes! That doesn't make any sense. How much time the resident spends grossing in a case has absolutely no bearing on the patient's ultimate prognosis!

That is true only god knows what a person's real prognosis is. But a more thorough gross gives us a better estimate of the person's ultimate prognosis. If you spend five minutes and find five negative lymph nodes, and conclude pN0 in a pT2 case that has a much different predicted prognosis than if you spend 1 hour and find 50 lymph nodes with 1 being positive.

 

[Gene_]

That is true only god knows what a person's real prognosis is. But a more thorough gross gives us a better estimate of the person's ultimate prognosis. If you spend five minutes and find five negative lymph nodes, and conclude pN0 in a pT2 case that has a much different predicted prognosis than if you spend 1 hour and find 50 lymph nodes with 1 being positive.

No doubt the predicted prognosis will be different, but the question is how accurate that prediction really is. I believe the only accurate prognostic information that can be determined by exhaustive lymph node searches is "true" N0 status. What does 1 lymph node positive out of 50 really mean? Is that an N1 or something along the lines of micromets/itcs in breast?

 

[raspberry009]

We were taught that you were required to find and submit all the lymph nodes. 12 were only required for adequate staging.

There has actually been a study on this...
Nodal staging in colorectal cancer: should distant lymph nodes be recovered in sugical specimens? Hum Pathol; 2009;40 (April):552-557

Take home pearl, accurate LN staging should be performed <5 cm from tumor on either side.

 

[malchik]

The colon lymph node thing is the classic example of drawing the wrong conclusions. Someone has published a study which states that finding more lymph nodes in colon cancer cases improves prognosis (or something like that). The lesson that should be learned is not that you HAVE HAVE HAVE to find 12 nodes! That doesn't make any sense. How much time the resident spends grossing in a case has absolutely no bearing on the patient's ultimate prognosis!

Well said Yaah.
Another erroneous conclusion people are making: why do you equate number of lymph nodes recovered to thoroughness of the search, or that those with lower counts are being understaged? The data do not show that any understaging is happening or even that fewer counts equate with shorter time. In my experience it is actually the opposite. A more reasonable conclusion, especially if the colons under consideration were grossed in the same hospital/lab, is that the PATIENTS THAT DO BETTER ACTUALLY HAVE MORE LYMPH NODES in their pericolonic fat. Anti-tumor immune response perhaps?

I was treated like a rising star of GI pathology when I found 40 lymph nodes in a tiny specimen and yelled at when I had a huge colon with 10. It had nothing to do with my search! In both cases all the fat was entirely submitted.

 

[2121115]

I was treated like a rising star of GI pathology when I found 40 lymph nodes in a tiny specimen and yelled at when I had a huge colon with 10. It had nothing to do with my search! In both cases all the fat was entirely submitted.

I've had a similar experience. Usually they are either easy to find right off and you find a lot or they just aren't there and you are going to spend a lot of time looking and find very few. Not many in betweens in my experience.

All of the above comments are why I think there isn't good evidence to support that getting some arbitrary number of lymph nodes in these cases equates to good patient care, adequate staging or quality practice (PQRI, etc). Just do a quality dissection and submit all of the LN's you find. That should be plenty.

 

[green mantis]

we do 5 cassettes for the first 10 g, then one cassette for every 5 g thereafter. i've seen cancer lurking in there more than few times

I don't quite remember the rule that we used, but I know we were supposed to entirely submit if it fit in 8-10 cassettes.

I was treated like a rising star of GI pathology when I found 40 lymph nodes in a tiny specimen and yelled at when I had a huge colon with 10. It had nothing to do with my search! In both cases all the fat was entirely submitted.

This fact has always annoyed me. For some reason, people believe that the length of the segment equates w/ the number of lymph nodes that can be found.

Since we're on the topic, which fixatives do you use in your LN hunts? I've used Bouin's, Dissect-Aide & Carnoy's. Are there others that would be better? My hospital doesn't allow Dissect-Aide & Carnoy's.


----- Antony

 

[TMZ2007]

The colon lymph node thing is the classic example of drawing the wrong conclusions. Someone has published a study which states that finding more lymph nodes in colon cancer cases improves prognosis (or something like that). The lesson that should be learned is not that you HAVE HAVE HAVE to find 12 nodes! That doesn't make any sense. How much time the resident spends grossing in a case has absolutely no bearing on the patient's ultimate prognosis!

Yeah, but if you don't submit and evaluate 12 nodes, the surgeons get dinged as a "quality control measure" by somebody (or at least that's what they keep telling/yelling at us at GI tumor board).

Additionally, those same studies stating a correlation between prognosis and LN count are front and center in the AJCC staging manual, meaning the chance of any of this changing in the next 8 years or so is practically nil.

 

[yaah]

Since we're on the topic, which fixatives do you use in your LN hunts? I've used Bouin's, Dissect-Aide & Carnoy's. Are there others that would be better? My hospital doesn't allow Dissect-Aide & Carnoy's.

I always felt formalin worked the best. Bouins made the nodes bright but it firmed everything up so you couldn't feel them at all. Never had much success with other ones either although experience was limited. I always tried them after I had done a standard lymph node hunt and they never helped highlight any more. The formalin fixed stuff often did have the nodes get firmer and stand out though.

 

[2121115]

I think that regardless of what fixative you use, the nodes are either there or not and if a quality dissection is performed the ones that are there will be found. If they are there you'll usually find them relatively quickly right under the tumor. If you don't find 12 relatively quickly it is going to be a pretty long battle and the likelihood of it ending well goes down significantly. In the latter types of cases, it is almost quicker to just block in all of the fat up front and get it over with, since going back once is the strongest predictor of going back a 3rd and 4th time.

 

[malchik]

i think that regardless of what fixative you use, the nodes are either there or not and if a quality dissection is performed the ones that are there will be found. If they are there you'll usually find them relatively quickly right under the tumor. If you don't find 12 relatively quickly it is going to be a pretty long battle and the likelihood of it ending well goes down significantly. In the latter types of cases, it is almost quicker to just block in all of the fat up front and get it over with, since going back once is the strongest predictor of going back a 3rd and 4th time.

truth

 

[deleted314957]

and then comes the question of what is counted as a node and what is not counted as a node. this was covered in a very recent article. show the same slide to 10 paths and get counts of from something like 4 to 12? I guess if you would count it if it had a met in you should count if it didn't.

 

[KCShaw]

That's a new one to me. A lymph node which isn't a lymph node...?

On the fixative question, generally my specimens were either totally unfixed or not very well fixed, as we tended to gross them not long after surgery. A percentage would be opened at the end of the day and left for the next day, of course. Point being, my node searches were almost exclusively on fresh -> formalin fixed tissues only. The look and feel of LN vs fat vs fibrosis just becomes more clear over time, and you get better at squishing and confidently eliminating sections of tissue from further search. The only times I prodded through other fixatives, mainly some kind of dissect-aide stuff supposedly wiping out the fat for us, were because I was asked by someone to help out -- generally quite horrible experiences, IMO, because of the dissect-aide-like crap (not sure if that's exactly the fluid it was) changing the look and feel of everything. But if that's how you trained, then I'm sure the final outcomes are similar in comparison.

I agree that doing a "quality dissection" is what's needed, not a number per se, unfortunately what constitutes a quality dissection is evidently not that easy to establish. (Ergo, some people use a number in lieu of a more appropriate gauge, such as trusting the words and experience of the grosser.)

 

[djmd]

That's a new one to me. A lymph node which isn't a lymph node...?

The ol' how many lymphocytes do you need (or how big does it have to be before it counts as a lymph node)...

Some people will count a cluster of lymphocytes as a lymph node as long as it is "big enough" (with different people having different ideas what that big enough is)...

Others prefer the "has to have an identifiable capsule/sinus" rule.

 

[KCShaw]

Ah yes. I've always thought of lymphoid aggregates as...lymphoid aggregates. I don't recall 1 or 2 of them potentially a "significant" difference in my overall lymph node counts, but it's been a while. On the rare occasion tumor was associated with one, that was simply metastatic tumor (as I recall).

 

[zao275]

Others prefer the "has to have an identifiable capsule/sinus" rule.

That's how I do it. I don't care how small it is, as long as it has a capsule/sinusoid.


Regarding fixative, I have tried soaking the fat in 100% methanol or isopropanol, but I have been less than pleased with the results. If I am low on my node count, I usually keep a spare jar of lymph nodes that I can take a few from....oh, wait, um nevermind. Ok, just kidding (a surgeon or someone actually suggested this half jokingly once!). No, what I really do is throw in 10 or 20 extra cassettes of fat. It may be wasteful as far as histology costs go, but it saves a lot of time and often tiny nodes show up. If no nodes are there, well normal fat is easy to screen like lightning at 2x. I am not talking about doing this at first, but only after a thorough search fails to find many nodes (like usual, right?).

 

[raspberry009]

Since we're on the topic, which fixatives do you use in your LN hunts? I've used Bouin's, Dissect-Aide & Carnoy's. Are there others that would be better? My hospital doesn't allow Dissect-Aide & Carnoy's.


----- Antony[/QUOTE]

We do all our dissections fresh, for the few cases that I have done it on formalin fixed tissue... I much prefer it fresh, catch a wiff of those fumes and I'm sure I'm losin one of my few needed neurons by the minute...

 

[gbwillner]

Since we're on the topic, which fixatives do you use in your LN hunts? I've used Bouin's, Dissect-Aide & Carnoy's. Are there others that would be better? My hospital doesn't allow Dissect-Aide & Carnoy's.


----- Antony

At my institution people do it different ways. Some use dissect aid because it theoretically turns LNs white. I think its an absolute waste of time, and I'm also sure sniffing that stuff will kill you. As a first year, I tried it once or twice, and got high just standing around it.
I just make sure the fat is very well preserved in formalin. I then take serial sections, 2/3 mm, with a long blade, of all the fat. I ignore the epiploica- there is nothing in there. Most nodes are near the serosa if they are hard to find. I can only think of one or two times I was unable to find 12 nodes, even in the irradiated specimens.
One attending swears there is only one valid method- to systematically squeeze all the fat through your fingers or the back of the blade to "release" all the nodes. I used to do this, but it literally took 1 hr+ for any give colon, and with our volume that would be suicide.

 

[gschl1234]

Since we're on the topic, which fixatives do you use in your LN hunts? I've used Bouin's, Dissect-Aide & Carnoy's. Are there others that would be better? My hospital doesn't allow Dissect-Aide & Carnoy's.


----- Antony

We do all our dissections fresh, for the few cases that I have done it on formalin fixed tissue... I much prefer it fresh, catch a wiff of those fumes and I'm sure I'm losin one of my few needed neurons by the minute...[/QUOTE]

Out of curiosity, for GI specimens (colon, esophagogastrectomies, whipples) do you guys do staging on frozen? Specifically, I'm wondering about LN. What's the average turn-around time? How many people are involved in the grossing/freezing? Thanks.

 

[2121115]

Out of curiosity, for GI specimens (colon, esophagogastrectomies, whipples) do you guys do staging on frozen? Specifically, I'm wondering about LN. What's the average turn-around time? How many people are involved in the grossing/freezing? Thanks.

Freezing LN's on a colon for staging? WTF?

 

[green mantis]

Out of curiosity, for GI specimens (colon, esophagogastrectomies, whipples) do you guys do staging on frozen? Specifically, I'm wondering about LN. What's the average turn-around time? How many people are involved in the grossing/freezing? Thanks.

I've never heard of staging based on a frozen section, but I'm wondering if this is what you're asking about. I've had instances where I've done a frozen section on a liver biopsy on a patient w/ colon, pancreas, stomach, etc cancer. If it's positive, they sometimes stop the procedure because you've upstaged the patient, & they're no longer a candidate for surgery.


----- Antony

 

[gschl1234]

I've never heard of staging based on a frozen section, but I'm wondering if this is what you're asking about. I've had instances where I've done a frozen section on a liver biopsy on a patient w/ colon, pancreas, stomach, etc cancer. If it's positive, they sometimes stop the procedure because you've upstaged the patient, & they're no longer a candidate for surgery.


----- Antony

I'm asking about the Mayo specifically because I thought the original poster I quoted was a resident there. I heard they do "forzens on everything" and was wondering how far they took it (like complete staging, for example).

 

[KCShaw]

I had heard various things about that institution specifically with regard to frozens, including that cases would be finalized on frozen section, with tissue thawed but never paraffin embedded. Among other things I don't quite believe enough to repeat -- I'm fairly sure there were exaggerations in there somewhere. But certainly that they freeze essentially everything, and evidently have a lot of confidence with it. (So it would be interesting to hear how exotic they actually tread.)

 

[bluesdude]

... one of my pet peeves was residents announcing they were "done" when they had reached some given number, despite there being a significant amount of relevant yet completely unsearched tissue before them...

I agree. It's stupid to say such a thing out loud. One should keep that kind of thought process to themselves

 

[raspberry009]

I had heard various things about that institution specifically with regard to frozens, including that cases would be finalized on frozen section, with tissue thawed but never paraffin embedded. Among other things I don't quite believe enough to repeat -- I'm fairly sure there were exaggerations in there somewhere. But certainly that they freeze essentially everything, and evidently have a lot of confidence with it. (So it would be interesting to hear how exotic they actually tread.)

Point well taken. However, this is a good place to get the facts strait. Everything was at least blocked (paraffin embedded) and we would look at a section of tumor the following day do be sure the diagnosis was right. CAP gave Mayo an overhaul 2 years ago and said that everything needed to be looked at again the following day, so now that is what we do. Some consultants are confortable to sign out and stage most of the cases on frozen.

Patients come here because of the service we provide. People are now getting intraoperative radiation because of this quite frequently. When margins are positive on a breast case, the patient can get an additional resection and not have to wait for another operation. Honestly, if any of my relatives have cancer, I would bring them here.

Breast and melanoma cases where the sentinal node is negative are not ever signed out at frozen because of the chance of finding a micromet. But when you see tumor most of the time is tumor, very rarely is it not (which will get fixed the following day).

Remember, we do this every day and get really comfortable looking at things on the T. blue. Most consultants are aware of the limitations on cases and hold the case over to sign out the following day when all the H&Es come out. Yes, mistakes are made, but usually those will get corrected the following day and an amendment will be issued.

Everything is looked at on H&E the following day. EVERYTHING.

 

[gschl1234]

Some consultants are confortable to sign out and stage most of the cases on frozen.

That's really interesting. What about technically challenging frozens with lots of fat? I would think breast and lymph nodes would be the most difficult to get good sections of. Do they usually give a prelim frozen dx and hold for permanents if you can't get a complete cross-section?

 

[KCShaw]

Definitely not a common way of doing things, but were I in the general surg-path arena I'd certainly want to see that personally (rotation or somesuch). As with a lot of things, it comes down to what one is comfortable with and/or does on a regular basis. If an institution can back up their processes with data and success, great. Sounds like the frozen section Mecca.

 

[raspberry009]

That's really interesting. What about technically challenging frozens with lots of fat? I would think breast and lymph nodes would be the most difficult to get good sections of. Do they usually give a prelim frozen dx and hold for permanents if you can't get a complete cross-section?

Actually pancreas is the hardest to cut, oddly enough. With fatty specimens, we cut them to be a little smaller than usual. Because we all know that if you can't get a good section, the histology is crap. But you'd be amazed what the cutters can do.

 

[gschl1234]

Actually pancreas is the hardest to cut, oddly enough. With fatty specimens, we cut them to be a little smaller than usual. Because we all know that if you can't get a good section, the histology is crap. But you'd be amazed what the cutters can do.

Are you talking about the uncinate margin? I find you can still usually get the inked rim pretty well even if you're missing a lot of the fat. The tumor section should be really easy.

 

[yaah]

Mayo uses a different system for freezing and cutting than standard cryostats. So fat is not as problematic, as far as I know. Although no doubt experience plays a big role also.