Recent Wall Street Journal article about pathology

[WEBB PINKERTON]

http://online.wsj.com/article/SB100...336.html?KEYWORDS=what+if+the+doctor+is+wrong

"Thomas Feeley, vice president of medical operations at MD Anderson, says as many as 25% of patients who arrive at the center with diagnoses for certain cancers such as lymphoma may receive a different diagnosis."

Is MD Anderson wanting everyone to send them cases for 2nd opinions or what? That 25 percent number sounds like BS to me.

---

Members don't see this ad.

 

[pathstudent]

http://online.wsj.com/article/SB100...336.html?KEYWORDS=what+if+the+doctor+is+wrong

"Thomas Feeley, vice president of medical operations at MD Anderson, says as many as 25% of patients who arrive at the center with diagnoses for certain cancers such as lymphoma may receive a different diagnosis."

Is MD Anderson wanting everyone to send them cases for 2nd opinions or what? That 25 percent number sounds like BS to me.

It says that 25% "may" get a change in diagnosis. I suppose it could say 100% "may" get a change in diagnosis. And if it is changed how do we know MD Anderson is right unless there is some ancillary test that can verify the result (FISH, IHC, etc...) There are no gods.

And the patient they used for an illustration ended up having follicular lymphoma. You hardly need to go to MD Anderson to figure that one out. That is breat and butter pathology.

I just saw a case where Hopkins called a tumor one thing and MGH called it another completely different. Yes it is a hard case, but one of them has to be wrong and possibly both. The idea that there is a place that knows all the answers is really BS.

 

[LADoc00]

Yep. That is an interesting read. Im now considering a cut rate 2nd read advert campaign for HMO patients.

What do you guys think of:"According to a recent Wall Street Journal article, up to 25% of Pathology diagnoses are changed upon second review. Isn't your health and that of your loved ones worth a second look?"

[insert stock photo of serious science dude looking through microscope]

Quick, easy and inexpensive!
Pathology slide consult per case only $49.99
Complete chart review and written consultation only $99.99
In person consultation only $299.99/hour billed in 15 minute increments

????

Adverts online with local mags and newsprint stuff.

 

[WEBB PINKERTON]

I just hope CAP comes up with a response to this garbage. Remember CAPs response to the DCIS article that came out a few years ago? They basically threw many pathologists under the bus. They didnt make the public feel any better, but what do you expect from a field with so many semi-autistic people working in it. Gets my blood boiling just thinking about that. Cant wait to leave this field for good.

 

[2121115]

The patient used as the example never had a pathology diagnosis overturned. She was told she had ovarian cancer, without a tissue diagnosis, and went to MDA for treatment. When they actually did the biopsy she had follicular lymphoma. If they mean that 25% of clinical diagnoses are overruled by pathology then I would probably believe that.

Also just after quoting the 25% number, they state that in actuality it is 3% based on the data.

 

[KluverB]

Sh*tty article. I especially like the part where "Pathologists performed another biopsy"... and found this time it was not invasive cancer. LOL!

 

[KCShaw]

We preachin' to who, the choir? Have to find a way to reach the real target audience.

 

[Ruination]

Sh*tty article. I especially like the part where "Pathologists performed another biopsy"... and found this time it was not invasive cancer. LOL!

The best part it is that there was probably a small focus of invasion on the original and when they biopsied other areas of the cervix, voila- it was gone!

 

[Pathguy11]

http://online.wsj.com/article/SB100...336.html?KEYWORDS=what+if+the+doctor+is+wrong

"Thomas Feeley, vice president of medical operations at MD Anderson, says as many as 25% of patients who arrive at the center with diagnoses for certain cancers such as lymphoma may receive a different diagnosis."

Is MD Anderson wanting everyone to send them cases for 2nd opinions or what? That 25 percent number sounds like BS to me.

I don't see this article as that incorrect...perhaps misleading due to media spin but not inaccurate. I actually do see this quite often with lymphoma classification at MD Anderson. A lymphoma is called one thing at an outside community based hospital for example, sometimes based on limited testing resources, and then gets further or reclassified at the large cancer center with a thorough workup. Here is another example I have seen quite often where the outside interventional radiologist gets a sample that is read out in the repor as "minute fragments of garbage without definitive diagnosis". I personally reviewed many cases like this when the patient come for treatment to the cancer center. I must admit that the interventional guys at big cancer centers like MD Anderson are pretty phenomenal about getting adequate samples and have dedicated large Cytopathology departments that do immediate assessments on ALL cases. Typically an IR guided FNA case will results in smears, cell block and usually good cores. In other words ample specimen for a diagnosis, whereas many times the IR's at smaller places just don't have the same success or Cytology support to ensure a great sample. 3% is currently the number I have heard used for ALL cases that come to MD Anderson for review that result in a "significant" change in diagnosis. Most cases however are confirmed to be correct from the outside report. These numbers of discordant diagnoses are much higher in the hematologic sections mainly because many outside reports give a vague diagnosis. And you are correct in saying there are no all knowing gods of Pathology at any institution, but the Pathologists at places like MD Anderson see a hell of a lot more cases in their single area of expertise in one year than most general Pathologists would in their lifetime. They can also afford to do the most thorough and cutting edge ancillary studies to get the needed diagnosis, not to mention there are tons of clinical protocols for subsets of patients with every molecular mutation of a tumor that you can think of. So though the article is misleading in order to get attention, I think the quote was taken a bit out of context... or at least without background explanation.

Pathguy11

 

[2121115]

I don't see this article as that incorrect...perhaps misleading due to media spin but not inaccurate. I actually do see this quite often with lymphoma classification at MD Anderson. A lymphoma is called one thing at an outside community based hospital for example, sometimes based on limited testing resources, and then gets further or reclassified at the large cancer center with a thorough workup. Here is another example I have seen quite often where the outside interventional radiologist gets a sample that is read out in the repor as "minute fragments of garbage without definitive diagnosis". I personally reviewed many cases like this when the patient come for treatment to the cancer center. I must admit that the interventional guys at big cancer centers like MD Anderson are pretty phenomenal about getting adequate samples and have dedicated large Cytopathology departments that do immediate assessments on ALL cases. Typically an IR guided FNA case will results in smears, cell block and usually good cores. In other words ample specimen for a diagnosis, whereas many times the IR's at smaller places just don't have the same success or Cytology support to ensure a great sample. 3% is currently the number I have heard used for ALL cases that come to MD Anderson for review that result in a "significant" change in diagnosis. Most cases however are confirmed to be correct from the outside report. These numbers of discordant diagnoses are much higher in the hematologic sections mainly because many outside reports give a vague diagnosis. And you are correct in saying there are no all knowing gods of Pathology at any institution, but the Pathologists at places like MD Anderson see a hell of a lot more cases in their single area of expertise in one year than most general Pathologists would in their lifetime. They can also afford to do the most thorough and cutting edge ancillary studies to get the needed diagnosis, not to mention there are tons of clinical protocols for subsets of patients with every molecular mutation of a tumor that you can think of. So though the article is misleading in order to get attention, I think the quote was taken a bit out of context... or at least without background explanation.

Pathguy11

MDA also has the same situation that other academic institutions have... which is that they can call it whatever they want and the case then gets put away and no one will ever pull it to send somewhere else for another opinion. In that situation you can make borderline calls and be more aggressive. In a community practice you have to sign things out in such a manner that no one else in the entire country would disagree, because if it gets sent out (and it probably will, and there is no way knowing to whom) if anyone disagrees then the community pathologist is always the one that is wrong.

How many of MDA's diagnoses get changed when their cases get sent out? I bet it is close to 3%. That is the nature of the game.

 

[lipomas]

I don't think it's total garbage, it's just semantics. You could review outside slides and see something minor that you report that "changes the diagnosis" but doesn't really change the clinical significance. Like changing a prostate core from 3+4 to 4+3.

The problem is, and I have a problem with a lot of this proliferating literature on "expert review changes the diagnosis" stuff, is that how do we know the "experts" are right? There is a new article in AJSP by Epstein which states essentially that "outside pathologists miss lots of Gleason pattern 5." http://www.ncbi.nlm.nih.gov/pubmed/22035764
I read the article. There is I think one picture. It is an obvious pattern 5. But yet close to 50% of all second opinion cases with pattern 5 diagnosed at JHH had no pattern 5 reported by the outside pathologist. OK, my question is, "HOW DO YOU KNOW THAT YOU ARE RIGHT?" I mean, this is essentially one pathologist disagreeing with dozens of others and saying, "I am right, you are wrong." Are you sure? The article does not acknowledge this at all, nor does it acknowledge that Gleason Grading can be subjective.

Or it's like endometrial hyperplasia. This is a good one here: http://www.ncbi.nlm.nih.gov/pubmed/16400640
Experts disagree with contributing pathologists, but disagree with each other just as much. Who's right? Clearly the experts are!

The follow up to this article should say that "cases diagnosed at MDA sent elsewhere had a disagreement 25% of the time." That might cause the media's collective heads to explode.

 

[mlw03]

because you're Dr. Epstein, and in GU pathology your word is the final one. agree it doesn't mean the "experts" always right, but my point is that when you go to an academic medical center and have your material reviewed, in theory, it's by someone with more expertise than a community pathologist. now, i wholly agree that's only in theory and that there are plenty of folks working in academic medical center with no more or less expertise than those in the community.

I don't think it's total garbage, it's just semantics. You could review outside slides and see something minor that you report that "changes the diagnosis" but doesn't really change the clinical significance. Like changing a prostate core from 3+4 to 4+3.

The problem is, and I have a problem with a lot of this proliferating literature on "expert review changes the diagnosis" stuff, is that how do we know the "experts" are right? There is a new article in AJSP by Epstein which states essentially that "outside pathologists miss lots of Gleason pattern 5." http://www.ncbi.nlm.nih.gov/pubmed/22035764
I read the article. There is I think one picture. It is an obvious pattern 5. But yet close to 50% of all second opinion cases with pattern 5 diagnosed at JHH had no pattern 5 reported by the outside pathologist. OK, my question is, "HOW DO YOU KNOW THAT YOU ARE RIGHT?" I mean, this is essentially one pathologist disagreeing with dozens of others and saying, "I am right, you are wrong." Are you sure? The article does not acknowledge this at all, nor does it acknowledge that Gleason Grading can be subjective.

Or it's like endometrial hyperplasia. This is a good one here: http://www.ncbi.nlm.nih.gov/pubmed/16400640
Experts disagree with contributing pathologists, but disagree with each other just as much. Who's right? Clearly the experts are!

The follow up to this article should say that "cases diagnosed at MDA sent elsewhere had a disagreement 25% of the time." That might cause the media's collective heads to explode.

 

[WEBB PINKERTON]

Articles like this play right into the hands of the specialty labs. Make the local pathologist look like he/she is incompetent and the specimens need to be seen by an "expert". Bostwick (or whomever) sales rep might as well start taking it with them when they try to steal business. It will lead in well to their "lab developed tests" that may or may not be completely useless.

 

[mlw03]

so is there evidence that a pod lab GU pathologist does a better job than a community pathologist? if not, what do their sales reps claim when they make their pitch?

 

[2121115]

so is there evidence that a pod lab GU pathologist does a better job than a community pathologist? if not, what do their sales reps claim when they make their pitch?

There is no evidence that any type of pathologist does better than any other type of pathologist. It is all marketing. Even at academic centers.

 

[Pathguy11]

MDA also has the same situation that other academic institutions have... which is that they can call it whatever they want and the case then gets put away and no one will ever pull it to send somewhere else for another opinion. In that situation you can make borderline calls and be more aggressive. In a community practice you have to sign things out in such a manner that no one else in the entire country would disagree, because if it gets sent out (and it probably will, and there is no way knowing to whom) if anyone disagrees then the community pathologist is always the one that is wrong.

How many of MDA's diagnoses get changed when their cases get sent out? I bet it is close to 3%. That is the nature of the game.

I definitely can't say I disagree with your point.

Pathguy11

 

[yaah]

because you're Dr. Epstein, and in GU pathology your word is the final one. agree it doesn't mean the "experts" always right, but my point is that when you go to an academic medical center and have your material reviewed, in theory, it's by someone with more expertise than a community pathologist. now, i wholly agree that's only in theory and that there are plenty of folks working in academic medical center with no more or less expertise than those in the community.

I saw that paper too - I was a little perturbed that there was no acknowledgement that this paper seemed to be almost entirely based on one person's opinion. That person's opinion is obviously well informed, but the paper was treating it like gospel truth. Further evidence to this is that there have been multiple papers in the GU literature recently about the level of agreement amongst "GU pathologists" on controversial topics like extraprostatic extension, micropapillary differentiation, etc. THere is far from 100% agreement.

IMHO this is more evidence of the spread of "eminence - based pathology" at the expense of "evidence-based" pathology. A lot of what we do is opinion based but we can't all read minds, there have to be some objective criteria.

 

[smellycat]

I like how they were going to treat ovarian cancer without a biopsy. Oh and the biopsy when finally done properly was FL. Yes, because ovarian CA looks like FL.....

 

[member0007]

i read the article too... it was a totally entertaining read. the only thing of value to me was one of the lists on the side - "changing the diagnosis". the other list "getting a second opinion" was for the primary care physicians and how they screw up... i don't get why that was part of this article, other than to mislead people into thinking pathologist are calling something viral and then changing it to cancer, haha.

 

[lipomas]

I wonder how many of these 25% are patients with sarcomas who are getting some fibromyxodermatosclerosarcoma renamed myxofibroscleroxanthosarcoma or something like that. No change in treatment, but maybe based on case reports of 5 cases myxofibroscleroxanthosarcomas recur at a rate of 45% as opposed to 30%. So that's significant!

 

[yaah]

I wonder how many of these 25% are patients with sarcomas who are getting some fibromyxodermatosclerosarcoma renamed myxofibroscleroxanthosarcoma or something like that. No change in treatment, but maybe based on case reports of 5 cases myxofibroscleroxanthosarcomas recur at a rate of 45% as opposed to 30%. So that's significant!

Yeah, or they are patients who had some soft tissue lump excised 3 years ago and now it has recurred, and with the benefit of hindsight it can be called something other than the original diagnosis.

Hindsight changes a lot of diagnoses!

 

[Path2009]

I thought I would resurrect this post. One of the things that bothered me most about the WSJ article was that the journalist ignored published data. We have an article coming out in Archives that has solid data that differs drastically from the 26% quoted by the MD Anderson representative.

Here is the link: http://bit.ly/MitBgu

Here's the executive summary:

From 2005-2010, we reviewed the pathology of nearly all patients that were referred to Mayo Clinic for treatment. We had major disagreements only 0.6% of the time (457 of 71,811 cases). The most frequent areas of disagreement were gastrointestinal (80 cases; 17.5%), lymph node (73;16.0%), bone/soft tissue (47; 10.3%), and genitourinary(43; 9.4%). Further, we freely admit that our disagreeing diagnosis was not always the correct one. In a subset of these cases (n=86 disagreements from July 2009 – Dec 2010), subsequent tissue showed that the original diagnosis was correct 15.1% of the time (13 of 86 cases).

In conclusion, community pathologists are excellent, and get it right most of the time. Academic pathologists are not always right.

 

[NuckingFuts]

so is there evidence that a pod lab GU pathologist does a better job than a community pathologist? if not, what do their sales reps claim when they make their pitch?

A recently-published study by Rajal Shah (who curiously now heads a private practice [pod lab?] in Texas) and friends just put out this gem:

Variant (divergent) histologic differentiation in urothelial carcinoma is under-recognized in community practice: Impact of mandatory central pathology review at a large referral hospital.
Urol Oncol. 2012 May 17. [Epub ahead of print]

 

[path19]

It's too bad that the Mayo study is in Archives which is never read by clinicians whereas urologists are now going to see the Shah study.

 

[Path2009]

I agree. I think its great to publish in clinician journals.

 

[NuckingFuts]

The thing of it is, if a physician signs out a limited type of specimens day in day out (subspecialist), as a matter of course, that person will certainly make diagnoses that would seem mundane to her, while at the same time the same diagnosis would be a rarity to a generalist. If you don't think of it, you will never diagnose it.

That's why the buck stops with people like Epstein, Fletcher, Rosen, Amin, etc.

The criticism I've heard from generalists in regards to subspecialists is that they (the generalists) have sometimes regarded the subspecialists as "1-trick ponies." I suppose this analogy may hold water in some circles, but tell you what- them 1-trick ponies are the ones "telling it like it is" when it comes to the "final say."

I'm not making an argument for what I consider to be right or wrong, I'm simply making an observation from the way things seem to be playing out in our field.

How much longer do you guys see pathology as a field existing the way it has over the years?

With continually growing pressure from the clinical component of medicine and even patients nowadays to have their specimens interpreted by "dedicated subspecialists", I often wonder if the vast majority of pathology, at least in relatively populated areas, will house any generalists in the next 20 years or so.

And will pathologists still be referred to as such? I personally think we ought to be called diagnostic specialists- we've come a long way from the days of Bloodgood.

 

[2121115]

The thing of it is, if a physician signs out a limited type of specimens day in day out (subspecialist), as a matter of course, that person will certainly make diagnoses that would seem mundane to her, while at the same time the same diagnosis would be a rarity to a generalist. If you don't think of it, you will never diagnose it.

That's why the buck stops with people like Epstein, Fletcher, Rosen, Amin, etc.

The criticism I've heard from generalists in regards to subspecialists is that they (the generalists) have sometimes regarded the subspecialists as "1-trick ponies." I suppose this analogy may hold water in some circles, but tell you what- them 1-trick ponies are the ones "telling it like it is" when it comes to the "final say".

The issue is that those cases only come around once every blue moon. A good general pathologist can sign out almost anything, and even more when sharing the case with a competent partner.

 

[Path2009]

The issue is that those cases only come around once every blue moon. A good general pathologist can sign out almost anything, and even more when sharing the case with a competent partner.

I agree - and I think our data supports that as well.

 

[lipomas]

A recently-published study by Rajal Shah (who curiously now heads a private practice [pod lab?] in Texas) and friends just put out this gem:

Variant (divergent) histologic differentiation in urothelial carcinoma is under-recognized in community practice: Impact of mandatory central pathology review at a large referral hospital.
Urol Oncol. 2012 May 17. [Epub ahead of print]

lol, I love the logic in the literature over this topic.

First papers were that divergent differentiation was more common in high grade tumors and maybe meant the tumor was worse

Then more papers said that we need to mention it because it's associated with worse tumors

Now we get a paper that says the community doesn't recognize it

And I just saw a paper that said even if it's there it doesn't matter because there is no independent prognostic ability of divergent differentiation.

So, therefore I conclude that it doesn't matter but we need to mention it because academics do. Now my bladder templates have a separate line for this! Typical tumor board conversation:
me: This tumor has focal squamous change
Clinician: What does that mean?
Me: That it's more likely to be a bad tumor
Clinician: But it's already in the nodes anyway, we know that it's a bad tumor.
Me: It might have some impact on therapy, supposedly
Clinician: Have they done studies on it?
Me: Yes, doesn't seem to impact therapy.
Clinician: Why are we reporting this again?