A Quality Pathology Report, Voice in the Wilderness

| July 27, 2008

What does the social networking MD network, Sermo, think about image enhanced reports, telepathology, externalizing pathology images, etc…?

This was a post on Sermo several months ago.  I thought of it recently when our group was asked to upload images into the hospital PACS for viewing by clinicians.  This would include both gross and microscopic images.  The institution provided excellent support in terms of software, upload application, training and documentation tools without a requirement to do so.  It can be done ad hoc as individual pathologists see fit for particular cases. 

For several years with different LIS and PACS systems I have made this part of my practice.  I found it useful for documentation, substantiating diagnosis, providing education to clincians and since patients may and do see their pathology reports, illustrations that may be informative if they can be explained or legends are complete.  Examples of the latter include illustrating mitoses or margins for documentating grading and staging when discussion with oncologist takes place. 

The problems, questions and concerns associated with this are numerous: what fields to select, what magnification, how many, value added, legal culpability if some action is or is not taken based on  image(s) rather than text diagnosis, cost, time and reimbursement.  Most of my colleagues always question the value added and how much effort it will add to their workload for what gain to their practice, division, department or institution.  Some get concerned that clinicians will copy images of the PACS to be used in papers and publications without due credit.  I have been doing this for years with radiology images and I have never had a radiologist mention a word.  I have even inquired if they would like to contribute and they inevitably refer me to the report and images on the PACS.  And what is the value added of uploading images of a basal cell carcinoma for a clinically suspected one or a benign diagnosis that is compatible with the clinical picture of a dermatology lesion.  I select dermatopathology cases because they seem to have the most say among clinicians whether they need it, want it, would use it, value added, legal implications since pathology is on the dermatology boards, etc…, much like discussions that take place for adding microscopic descriptions.  Can clinicians make out the microscopic description?  It is part of some if not all pathology reports depending on practice and the pathologists.  No doubt anatomic pathologists, like all specialties, like all fields, have their own personalities, concerns and "best practices".  This varies widely among anatomic pathology just in terms of number of blocks submitted, number of levels, special stains used, immunohistochemistry panels/reagents/kits, etc… and how those pathologists and practices work with clincians that utilize their services.  No 2 are alike. 

I think this post hits on this and discusses the issues amongst pathologists and clincians.  I scrubbed the Sermo IDs; most are pathologists, some are clinicians, I think you can tell which are which and get a sense of where Sermo pathologists responding to this think about these issues.

Would welcome your comments.



Here is what I believe is "quality pathology reporting"

1 . Micrsocroscopic description , describing salient features as to how the pathologist derived at the final diagnostic interpretation.

2 . Representative digital image(s) highlighting the salient features in the diagnosis , preferably a scanning or low power, if presentable, and/or at least an intermediate or high power view, demonstrating supportive architectural and cellular details ( this way other clinicians, GI/Derm, or pathologists may be able to review report).

3 . Availability of pathologist to show diagnostic images over the internet ( secured channel , no patient identifying information disclosed ), for immedate second opinion or review by other experts .

4 . Communication between clinician and pathologist with documentation of issues discussed on final report ( may be in a comment section ), i.e . the diagnosis of malignancy was discussed with the clinician at date and time, confirming the diagnostic impression.

We here use a high resolution digtal photomicroscope, which can easily be transferred online with still photograph or through screaming videography.

Our physicians here aprreciate this commitment to quality.  Would like to know your opinion in this regard.



I love it. Sounds exactly like what I’d prefer when reviewing a pathology report. (Especially the clinician-pathologist collaboration part)

I dont’ think this is about small or large practices being eliminated, because the technology currently is affordable; if your kid can afford a cellphone to textmessage, and iphone, then your pathology department surely can afford technology that would contribute to high quality pathology practice; just my opinion

those excellent pathologists you speak of should not have a problem with others reviewing their reports. SHOW THE WORLD YOUR DIAGNOSES; DON’T HIDE THEM IN A FILE, LIKE WE HAVE DONE SINCE THE BEGINNING OF TIME, and require FEDEX and UPS to deliver slides for second opinion, of course this will still take place.

Showing the photos is a natural extension of this once the technology becomes good enough, I guess. Unlike radiology, though, what’s in the photos is what we select, not the whole lesion (unless it’s tiny). So there’s still opportunity for bias, but it certainly would shine more sunlight on the case.

Here’s a question that maybe the nonpathologists can weigh in on too — I think with the advent of PACS on the wards, we all look at more radiology than we used to (and maybe catch things the radiologist misses from time to time). Would that happen with pathology, if microscopic images were available?

I doubt it. There isn’t the exposure to path like there is for radiology.

I would love it, but the labs I’ve dealt with seem too busy cranking out the specimens to do this kind of in-depth interaction. Probably like everyone else, they have to run harder just to stay in place income-wise.

Some clinicians can read xray and CT images just as well as a radiologist; not the case for pathology; however, the idea here is referenced to evidence-based medicine with regards to putting a digital image on the report, that could at least reflect how the primary reviewing pathologist came up with the diagnosis as rendered; ie malignant ductal cells to call breast cancer in a pediatric patient, or bizarre lymphoid cells to call a high-grade lymphoma in a young teen; this way other reviewing pathologists be it for the patient or clinician can say, yea I can see how the diagnosis came to be rather than, TRUST ME I’M YOUR PATHOLOGIST. Well, as I have stated before, tell that to the mother of a teen fighting for her life because the pathologist called an immature, malignant teratoma, a DERMOID CYST, and delayed treatment for nearly six months.

What I usually get for a micro is "microscopic examination substantiates the above diagnosis." If I get a micro it is because the pathologist has no idea what he/she is looking at. If I had it my way I would get a copy of the slide with the report.

These seem like good ideas, but I think that the images that would be added to reports would be the same that are shown at Tumor Board. If the image is to be placed in the report, then it will end up in the patient record. This would be okay if the entire case could be placed there, but with only a solitary (or maybe 2) images to show the area representative of the diagnosis, there may not be all the factors present in the image that is in the microscopic description. The legal beagles would have ammunition then to ask where in the image is the stuff mentioned in the microscopic. The answer of ‘it is only a representative image’ while acceptable to us, would just give the impression of the microscopic being ‘made up’ in a courtroom with a good lawyer. (Not that our cases would end up being sued, but with the catchall lawsuits out there nowadays, everyone gets named and there is no reason to give ammunition to anyone.) IMHO

That has been the oldest argument,and it does not stand up, because the slides and blocks, according to CAP accreditation standards, will be maintained for review for an appropriate period of time. We are simply suggesting that representative images of the most salient diagnostic feature (not all diagnostic components) should be digitally placed on the report for electronic and permanent record filing. For instance, if you make a diagnosis of malignancy, which is a microscopic diagnosis, then show features of malignancy (i don’t care about the final histological classification which can be confirmed on review of the entire slide for confirmation prior to treatment), you simply should not be able to justify a person’s breast coming off with ‘WORDS’ only, and no other evidence-based source. Let’s get modern; I am not affraid of the courts, when I use sound scientific based judgement to finalize my pathological diagnosis.

As per above, for instance with breast cancer, I include 1) picture of infiltrating malignant ducts, either in sheets, or degree of tubular formation; 2) higher power view of the nuclear morphology and, an example of mitosis; 3) the highest grade of DCIS, 4) angiolymphatic invasion and 5) lymph node status, with perinodal extension if present. A total of five pictures that generally take me 5 minutes to place on the report. I think this is representative of what the surgeon or oncologist needs to know to appropriately prognose and treat this patient.

we have a system in place that once you take the picture, the picture is automatically sent to the final pathology report and place in the appropriate position on the report.

It seems to me your immature teratoma example raises one problem that your system leaves unsolved (and no system other than peer review can solve satisfactorily). If the first pathologist only includes images of areas consistent with dermoid cyst and doesn’t see, recognize or photograph the immature areas, then the images in the final report will be consistent with dermoid cyst, not immature teratoma. It’s likely that a pathologist including images in her/his report will select those images from the fields s/he views as most diagnostic of her/his diagnosis.

Thus, images may help if the diagnosis is flat-out wrong/incomplete, but might not help in matters of grading, subclassifying, or finding malignant areas in a benign lesion. You only see what you see.

There are other clues in the final pathology report; like gross examination, and clinical correlation. First the gross showed a 14 cm necrotic tumor, and secondly there was a clinical suspicion that raised the possibility of malignant through xray studies, etc, that’s why a frozen section was performed. But your premise is correct that it requires a complete team to assess accuracy in diagnostic reporting in any field be it pathology or radiology; that is why the clinicians have a certain duty or responsibility to challenge the pathological diagnosis when it does not correlate with the clinical impression.

I am of the mind that most of us medicine docts want the best for the patient; having said this, I don’t think one, pathologist, would make up or cover up mistakes by putting normal or benign tissue on that which is obviously malignant; by the way the slide was not subtle with diagnostic features of malignancy; this was gross negligence. I want to emphasize to you in your youth, that we pathologists should never operate in a bubble.

Oh, by the way, if you know of incompetent pathologist that would do this, you can believe one thing, (s)he won’t practice pathology for long; the clinical colleagues will assure this, if not the pathology community; it is a tough specialty to practice, reed, and mistakes are costly in this profession, so I hope you don’t think it will be easy to make mistakes as you have aluded to above, without very serious consequence to your pathology career.

That’s why in my previous comment, I advise you that when in doubt, rule IT out ; this means, when you are not sure, YOU BETTER ASK SOMEONE. Get a second opinion, or send it to an expert with a lot more experience on such pathology issues, i.e questionable whether or not IT IS MALIGNANT.

Always remember this, you can be wrong on benign lesions (SK, fibroadenoma, demoid cyst, etc), you may apppear as an unwise pathologist among our colleagues, but there is no harm done, because benign lesions DON’T KILL; however, if you are wrong about malignant diagnoses, either way, you are screwed; why? because malignancy don’t go away, and eventually, if the patient is not treated and even sometimes if they are treated, malignancy WILL CAUSE DEATH TO THE PATIENT, (apart of the definition of malignant tumor)

This would be a welcome upgrade to my current service…very nice!

thanks; we just need to be persistent about bringing it into being; I think there is a push by the pathology community at large to at least include digital images on final pathology reports; hopefully, that would come to past, with eventually including a copy of the slide that would include representation of all diagnostic features to be sent to patient’s clinical doctor along with the final report, if the clinician requests it at no fee to patient or doctor.

We need to be reminded that patients go all over the country for treatment, and so this would be important, in addition to monitoring quality for those who have exclusive contracts, and lack external peer review.

This kind of information sharing is one of the obvious benefits of EMR technology and where I hope we are all heading. The transition is difficult, but really quite exciting. Keep it up….I love to see this stuff!
How much do you think telepathology would aid in this process? I gather that images that are sent out for telepath consult would be included in the final report along with the statement about peer review (or 2nd opinion or whatever statement is used by your service). Does this sound feasable to you?

I think, albeit not accepted by the general community of pathologists, telepathology will become the standard way of doing pathology practice, particularly for very remote location, like Wyoming. Here is what I envision how telepathology consultation service could work: 1) The telepathology consultant (expert or second reviewer) is first notified by the original viewing pathologist, and a code, series of alphabet and numbers similar to a surgical pathology number (unique identifyer) is given to the consultant for future reference (i.e. slide or case identification number that also identifies the patient) which is shown to the consultant at the same time as the slide is being viewed by the consultant;2) the consultant could copy unique images from the slide, selecting representative unique diagnostic features, and adding text comments as to what was discussed and the final conclusion, ALL TO BE SAVED ON A ELECTRONIC DATABASE, at selected time interval or ad infinitum for future reference.

Example of above:
Code, unique identifying number: TP080001, identifiying the patient as "Suzie Que"
Original Pathologist: Contributor
Telepathology Consultant(s): Dr. Expert
The features of the presented slide, as demonstrated by (selected images) are compatible with GIST; suggest correlation with CD117, ….. (other stains, or clinical studies).
Notice I included more than one(1) telepathologist/consultant for the reviewing pathologists

Before this discussion closes, I wanted to comment about a schemata as to how telepathology can be used:
Telepathology Consultant (usually affiliated with academic setting with cream of the crop, dermatopathologist, cytopathologist, hematopathologist etc.)<———->image/videotography transferred through electronic, network database<——> field pathologist, solo or group (good generalist, good diagnostic skills, and know when to seek second opinion). Notice the double arrow to suggests reversible communication; analogous to a soldier out in the battle field calling on higher ranks/experience or expertise to give help or second opinion.

The clinicians we service wouldn’t have clue as to what they were looking at if we provided a picture of the lesion on the report, and I don’t believe they have any urge to learn a little microscopic pathology…that’s our job. If you want to brighten up your report, then do it, but it all costs money if you print paper reports that would otherwise be produced in black and white. The images also clutter up the electronic world of data memory.

I wouldn’t spend the time adding a picture(s) to the report, because our average clinician is in no position to judge my diagnosis. If I’m cited in a lawsuit, the plaintiff’s counsel will have a pathologist review the glass slide (which is one reason why we keep them, folks); I wouldn’t want to be cornered/limited by the photos I supplied by some sharp attorney.

I tend to focus much more on HOW the information is presented to the ordering clinician, with clear cut comments/recommendations. I’ve learned over the years what diagnoses will generate a phone call (and rightfully so); as such I try to clarify points of the diagnosis at the time the clinician reads the report. I tend to repeat key words like "benign" in the report, in order to avert a disaster of organ removal for a benign diagnosis that the surgeon has never heard of before. I’m probably the target of much snickering by the local docs behind closed doors for my apparent repetition, but I don’t care. The studies that have come out in the pathology literature in the last few years clearly indicate that many clinicians are pretty poor at interpreting a pathology report…adding a nice color shot of the lesion I don’t think will improve that.

No, but it looks cool!!!  And its better than nothing; other pathologists can review what you call cancer or not, that ‘s all for patient safety since patients are going all over the country for treatment, and not in little village AMERICA.; plus, electronic pathology reporting is going to replace your "black and white" printer or fax; doctors are on the go, and are not going to wait for a faxed report, when they can go to their internet site and pull off the patient report whereever they are, regardless if there is a digital image on there are not. Just my humble opinion.

I would ask how many pathologists would be willing to put their name on an "outside" case review report (as we call cases that are shipped in because the patient was diagnosed elsewhere but are recieiving treatment at our institution) based just on what was shown in a few digital pictures that were selected by the original pathology department? If the answer is no, then what is the point to add the pictures for other pathologists? Any case that will be reviewed had better be reviewed with ALL the available information (i.e., the whole set of slides), or the entire digitized slide sets (which is now possible due to the slide digitizers on the market). If nothing else, it allows the pathologist to verify the slide/case accession number prior to penning a new report. How would you verify that the picture(s) with a limited view of the slide(s) are what are really on that slide (i.e., no screw ups in plugging in the digital image into the wrong patient report)?

I am a dermatopathologist, and most of our dermatologists have specifically asked NOT to have a photomicrograph on the reports. Because dermatologists train in dermpath in their residency, they feel it holds them "liable" for the accuracy of the diagnosis.

For the dermatologist above who would like to get a slide with each report, we DO provide this service to many of our referring physicians (obviously for free). We cut an extra slide for each case and send it out to them. It is nice because if they call you to discuss a case, you can both look at a representative slide at the same time.

I would not have a problem putting my name on a consultation report based on telepathology images. The disclaimer would be included of course (The evaluation and findings noted are based on evaluation of representative digital images only. If there is concern for a more worrisome entity, slides will be required prior to further evaluation). Or something to that effect. It would not be alot different then when peer reviewing representative slides of a colon cancer. The key is to include the term representative (sections/slides/digital images/whole slide images). At least for me.

Category: Digital Pathology News, Electronic Medical Records, Reports, TeleHealth

Comments are closed.