Biopsy / sample submission histories

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 17:53:27 GMT

I suspect the hive mind has a much better idea of what is going on in the world than we mere humans!

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 10:18:39 GMT

Were not much different from ants then?

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 10:16:12 GMT

It's true. Our subconscious actually runs the whole show, it just finds it convenient to let our consciousness think it is in charge!

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 10:13:59 GMT

Ommmmmmmmmmmm......................................

Histiocytoma

Funnily sleeping on a difficult case (not literally or the slide would break or get a bit smelly ) does help with enlightenment sometimes - I am sure you find that?

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 10:06:45 GMT

Maybe meditation to clear the mind before arriving at a diagnosis ;)?

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 09:58:01 GMT

Yes I agree - however that's human nature - what comes out of it can be detrimental but also can be very constructive - I don't think that most of the advancements in science would have been discovered without external influences and bias? However a bit more self-awareness (this fits to all walks of life) would certainly be a good thing?

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 09:55:29 GMT

Tell me about it. Easy to miss a detail when you are trawling through huge pages of history. Re the isolation, however, bizarrely the influencing factor doesn't have to be anything related to the case, or even to veterinary medicine at all. As I understand it, the number of the bus you travelled to work on (as if!) can subconsciously affect your judgement.

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 09:52:20 GMT

Yes your right Alex - I think as we are isolated somewhat we don't have so many influencing factors so maybe it affects us less? As you say we have the ability to read the slides without the history and I think that's what we do or certainly question - however I still have to ask myself whether I am making the right diagnosis after I have had the third epitheliotropic lymphoma of the day

BTW Alex - I don't know of any studies off hand - However I was taking about a short and concise relevant histories not a full history. It's up to the vet to supply a paragraph or bulleted history related to the sample. The samples with 10-15 pages of printed Computer notes are not the best

The trouble with didactic teaching is that we are taught in black and white (as its very difficult not to) but veterinary life is a spectrum of grey Experience with up to date knowledge and self analysis (and sharing information/consultation) I feel is the best approach (sorry if that sounded cheesy).

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 09:47:12 GMT

Sounds like you are well aware of possible bias which is great. I wonder how many times we as clinicians in general though get influenced without realising it. For (hypothetical) example, we are trying to decide whether to treat a possibly cushingoid dog, we have taken history and exam and clin path into account and it now hinges on whether we consider the post acth cortisol of 700 (for example) sufficiently high to start trilostane. Would our decision be different if just prior to this we had priced up an operation that came to £500, or one that had come to £1000? It's quite possible it would!

Alex

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 09:28:49 GMT

LOL - It's all healthy discussion and one which is very relevant.Sorry if my post came across as aggressive - it was not meant too - its just a subject that I am obviously very passionate about

I understand what you are saying about bias. However, I can only speak for my self in this, is that I am concious that I am not influenced by certain things. I HAVE to come to my conclusions without bias as much as possible. That is why we generally sit back and ask ourselves numerous questions. When I ask my colleagues their opinions I often get a number but then come to my conclusions - sometimes the next day to allow my brain to come to its conclusions.

I get your concept of the DDx - this could be a bias but we don't really use it like that - we try and assimilate all the information we have not just making one fact or comment the most important. Biopsy changes first, filter the history to find the apparent relevant data and then see if it confirms or disproves the clinicians working diagnosis(es).

All I can say is that the most important info you can supply with your sample is the animals personal details, sample site, duration, lesion (s) description,clinical symptoms, additional relevant diagnostics and medical history - the DDx is last on the list but can still be helpful in answering questions such as - "this is not pemphigus, this is not a neoplasm...."

Sure I agree read the slide blind before you read the history is a good concept and one which I employ as you say but if you take the history away you also take away the human's ability to interpret and with it the advancement of the case. The trouble with medicine is its ability to change and also as its a bit chaotic and heterogenous most this fall into the grey area requiring experience and interpretation and a differential diagnosis list?

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 09:18:41 GMT

Hi Richard,

I wasn't implying you would consciously change the diagnosis to fit the history, apologies if it came across that. I am talking about subconscious bias, and it is far more prevalent than we realise. The book Predictably Irrational by Dan Ariel is a fascinating read, and I would love one day when I have time to research how subconscious bias affects us all in the veterinary field. It is established that it occurs in the medical and dental fields already. For example anchoring is a powerful subconscious effect. If people are asked if Gandhi was older or younger than 140 when he died, and then asked how old he actually was, they give a higher number than if first asked if he was older or younger than 40.

It may be that we all need to take these sort of factors into account to improve our diagnostic accuracy. It's certainly not meant as critcism of you personally Richard, or pathologists in general. Pathologists are just one of the few groups (along with maybe specialist radiographers) that it is possible to isolate the findings from the history.

Just meant to provoke discussion, don't red star me!

Alex

Re: Biopsy / sample submission histories

Richard Fox — Tue, 17 Jul 2012 08:30:25 GMT

Hi Alex. I am afraid not. It really is nothing like that. I do not make my diagnoses fit the history - that would be negligent. I read my biopsy in conjunction with the history with bit of healthy scepticism and then see if the history fits. If it does great, if it doesn't then I re-read the biopsy and if I still think the history doesn't fit I then ask the clinician more questions. Sometimes the history supplied suggests one thing but the biopsy does not. I would then come to the conclusion that the biopsy may not be representative.

I think unless you have trained in cytology and histopathology (oh yes I am a vet and have been in practice), it seems it may be difficult to get this concept. I asure you that what I am saying is an integral part of my job and I am sure that the vast majority of my colleagues do and would say the same thing. I don't think this is a concept that is up for discussion really - its a fundamental part of sample submission. I don't really think you can compare MRI/Rads (my wife is a human MRI radiographer) with cytology and histopath - there are very different however she also say the history is very important - in her department they will not accept a patient without a relevant history (minimum criteria set for each submission).

Without a history I am afraid I cannot interpret - I can give a morphologic diagnosis but that of no use to a clinician! I need a history in many instances to give an aetiologic diagnosis. I think this is a concept that may not be easily understood> I could give so many examples but I don't have the time to type them

Edit: The concept really is that you can come to a histologic diagnosis (called a morphologic diagnosis) such as a chronic hyperplastic perivascular dermatitis but without a relevant history an (human) interpretation cannot be made - if you told me that you had skin scraped, hair plucked, biochem and haem normal, cultured and tried AB's , flea control but to no avail I would then suggest it is likely either a food allergy or atopy - that's is what you are paying for - the pathologists experience in taking the changes seen on biopsy/cytology and coming to a conclusion about what is going on and what is causing it.

I would also suggest that a differential diagnosis is not often useful in coming to a diagnosis. It tells us what the clinicians train of thought but I would say in about 50% of the cases submitted is incorrect.

Re: Biopsy / sample submission histories

Alex Gough — Tue, 17 Jul 2012 08:13:06 GMT

Richard, do you know of any human studies comparing the accuracy of histopath read blind, or with limited information eg signalment and location, to those with a full history. I have to say, given the human ability for subconscious bias, I would echo David's concerns that reading a slide with knowledge of the suspected diagnosis could easily bias the answer. That said, I appreciate how context can be important. For example, if reading a head MRI where there is a huge amount of information, knowing that the dog had peripheral vestibular disease might make one concentrate focus on the middle ears rather than the forebrain, and subtle abnormalities in the forebrain might be more likely to be ignored as irrelevant/artefact. Maybe a solution would be two reports - the first a purely descriptive one read blind, the second an interpretative one after reviewing the history. It sounds like you may already do this informally.

Alex

Re: Biopsy / sample submission histories

Evelyn Barbour-Hill — Mon, 16 Jul 2012 22:03:59 GMT

[quote user="Richard Fox"]

Sorry Evelyn - Do you mean you find it weird that vets don't fill out the form?

[/quote]

I do.

Re: Biopsy / sample submission histories

Richard Fox — Mon, 16 Jul 2012 22:00:59 GMT

Sorry Evelyn - Do you mean you find it weird that vets don't fill out the form?

Re: Biopsy / sample submission histories

Evelyn Barbour-Hill — Mon, 16 Jul 2012 21:54:17 GMT

[quote user="Richard Fox"]and yes I think a lot of nurses fill out the forms[/quote]

I find that weird. What's wrong with people?

Re: Biopsy / sample submission histories

Richard Fox — Mon, 16 Jul 2012 21:50:57 GMT

I also agree very much with Thomas and Catherine - If you bother to supply a history then the pathologist should read it - and yes I think a lot of nurses fill out the forms and things either get left out or lost in translation sometimes.

Re: Biopsy / sample submission histories

Richard Fox — Mon, 16 Jul 2012 21:30:12 GMT

There lays the problem David - As you are not a cytologist or a histopathologist then you obviously are unaware of the many issues we have as diagnostitions. I have to assure you that at least a limited history is very helpful in coming to a diagnosis that would be useful to a clinician and also stand up in court if needs be. I am afraid I have to completely disagree with you. Cytology is a very different discipline to histopathology with its pro's and cons - and cytology, in my view, has been recently pushed as a cheap and very reliable method of diagnosis - which is is not. It is cheap, it is quick but it has a very limited range where is is actually diagnostic - Cytology, in my and many of my colleagues (Worldwide) is only really a guide to further tests i.e. in most instances leading you down one path or another i.e. treatment or to go for biopsy. It can be very diagnostic in certain circumstances - lymphoma, mast cell tumours, lipomas but there are whole host of areas is isn't.

Your statement "diagnosis especially given the poor agreement between histopathologists/cytologists" really boils down to the fact that cytology is in most instances, inferior, to histopathology. Of course it has its place as suggested above and at least 1/3 of my case load is cytology. This I am afraid is the same in human medicine - cytology is now rarely used (unless for fluid analysis etc) and diagnoses are not often made on cytology interpretations from all the uman EQA meetings and NHS bulletins I have been to and read?

I think if you spent a few weeks in our line of work (which may make you shudder at the thought you may change your mind. You may find if you did send in a little more detail (especially as we have close association with dermatologists where a full history is extremely important) you might get more back? I know from my point of view that vets who phone in and discuss the case with me often find it useful and we both go away having added something to the case and our own experience.

Edit: Tim put my thoughts above more succinctly and eloquently than I could

Re: Biopsy / sample submission histories

Tim Cheyne — Mon, 16 Jul 2012 20:14:16 GMT

I have always felt that, broadly, there are two ways of submitting a sample to a laboratory. The first is a request for a specific test, CBC, serum titre, faecal egg count, etc, that needs only an ID and the sender's contact details. The result goes back with no interpretation, the responsibility for which lies with the submitting clinician. The second and preferred way is basically a request for help in confirming or suggesting a diagnosis. This requires a succinct clinical history and possibly a suggested differential diagnosis. Despite appearances, pathologists are only human and do like to know what they are handling, also a good history can be a guide as to which tests need to be applied and the order of priority for their application so that at least a preliminary report and interpretation can be returned quickly, to the benefit of the clinician and patient.

Tim.

Re: Biopsy / sample submission histories

David Mills — Mon, 16 Jul 2012 19:48:30 GMT

[quote user="CatherineThomas"]

This tends to be more for blood sample results rather than biopsy though.

[/quote]

Yes, every SA lab I've ever used simply copies and pastes standard differentials for abnormal measurements - complete waste of the £10 + VAT extra most places charge, and a bit of an insult when you've provided a detailed hx which evidently has not been looked at.

Funnily enough, VLA were always an exception to this - v.g. personalised reports.

Re: Biopsy / sample submission histories

David Mills — Mon, 16 Jul 2012 19:43:33 GMT

[quote user="Richard Fox"]

You must be frustrated when your clients can't give you any decent history?

[/quote]

This is rather different though. Pathologists examine a limited number of tissues/cells whereas a hx can cover a much wider range of presentations. I agree narrowing down to a body system is helpful e.g. FNA from a spleen. Equally, I don't see how history in biopsy samples changes things - liver biopsy for instance - surely there will be a hx of liver dz for a biopsy to be taken.

I don't necessarily think that more hx leads to better his to/cyto diagnosis especially given the poor agreement between histopathologists/cytologists - I would have thought more rigorous multi-centred agreed criteria for diagnosis would have been more useful than a vets conjecture on some forms such as differential diagnoses (Idexx particularly irksome) - surely that's why I'm sending the sample in...what more is required than age, location, and a brief description of appearance?

Re: Biopsy / sample submission histories

CatherineThomas — Mon, 16 Jul 2012 19:26:22 GMT

I always try to fill in the history on my submission forms but I find it quite frustrating when the interpretation comes back saying: have you checked this? or are there any supporting clinical signs? Often things which I have written on the form. I feel I have wasted my time writing it down. This tends to be more for blood sample results rather than biopsy though.

Re: Biopsy / sample submission histories

Thomas Johnson — Mon, 16 Jul 2012 17:06:11 GMT

I suspect part of the problem is that some vets take a sample, hand it to a nurse and say, "send this to the lab", and don't take the time to fill the form in themselves. Some nurses will ask the vet what to put in the history, others won't, or the vet has disappeared out of the building when the nurse fills in the form after doing their other jobs.

Re: Biopsy / sample submission histories

Richard Fox — Mon, 16 Jul 2012 16:27:45 GMT

I often look at the slide blind initially but require a history to confirm site and context. How do we know where the biopsy if from, what of, how long, treatment and response to. I think our comments would be huge with numerous Ddx's. I have to say histopath and particularly cytology (god only knows where the sample comes from without a history) is not a test that should be interpreted without a half decent report. If this were the case you would get far more non-definitive diagnosis!

Edit: BTW if you don't believe me have a chat with a human histopathologist - they have to dot every I and cross every T before they have even started! Go along to a human Histopath EQA meeting - its a real eye opener!

You must be frustrated when your clients can't give you any decent history?

Re: Biopsy / sample submission histories

David Mills — Mon, 16 Jul 2012 16:22:15 GMT

I do wonder about this sometimes. Surely it is sometimes far better for something as focused as a pathology sample for the histopath/cytology to be done 'blind' and therefore surely more objectively and nothing is missed? That goes for anything, of course - X-rays are particularly good for this, similarly echo exams, etc.