SNAP test grey zones

RE: SNAP test grey zones

Rob Loxley — Wed, 18 Nov 2015 11:25:53 GMT

[quote user="Andrew Kent"]

[quote user="Francisco Gomez"]

Did find a fiv+/leuk- in a SpeedDuo. Sent to external lab to find fiv-/leuk+.

CP changed their test to this brand to save costs, unfortunately theyre spending more due to the high sensitivity (or number of FP) in the new tests bcause the policy is sending those to an external lab to confirm...

[/quote]

I agree it is not a perfect test and agree that any positive should be validated externally or by other methods, but as a widely available screen its pretty useful.

[/quote]

I agree about preferring the IDEXX Snap to the Speed-duo FIV/FELV tests as far as readability goes.
There's a difference between being able to read what the inhouse test says and needing to send off to confirm a positive inhouse test

RE: SNAP test grey zones

Andrew Kent — Wed, 18 Nov 2015 10:58:09 GMT

[quote user="Robin Grimmer"]

[quote user="Andrew Kent"]

The only SNAP test I use is the FIV/FeLV.

The majority of the other snap tests are, at best, potentially misleading and at worst down right dangerous.

Andy

[/quote]

The canine parvo virus antigen one is pretty reliable, no?

[/quote]

Ok, yes, fair point I hadn't thought about enteric tests - I guess there is often some urgency about the diagnosis of parvo.

The Giardia antigen test is also reliable, but I am never in a hurry to diagnose Giardia so don't use it.

[quote user="Francisco Gomez"]

Did find a fiv+/leuk- in a SpeedDuo. Sent to external lab to find fiv-/leuk+.

[/quote]

I agree it is not a perfect test and agree that any positive should be validated externally or by other methods, but as a widely available screen its pretty useful.

Andy

RE: SNAP test grey zones

Anthony Dennison — Wed, 18 Nov 2015 10:25:39 GMT

[quote user="David Mills"]There's another in the pipeline about diagnosing Chlamydophila on conjunctival cytology too if you fancy poking some fluey cats in their eyes with a cotton bud...[/quote]

Well after GMAH sent over a couple of fluey cats that stopped intake of new animals for about 8 weeks, they're very fussy about what they accept now! But yes, if I get anything fluey I'll cotton bud it. Any particular stain?

I'm not fantastic at scanning hearts but can identify a thickened ventricle (I'd like to think). Only got a basic scanner so no doppler or other fancy modes, but I'll appreciate the practice!

RE: SNAP test grey zones

Anthony Dennison — Wed, 18 Nov 2015 10:06:36 GMT

[quote user="David Mills"]There's another in the pipeline about diagnosing Chlamydophila on conjunctival cytology too if you fancy poking some fluey cats in their eyes with a cotton bud...[/quote]

RE: SNAP test grey zones

David Mills — Wed, 18 Nov 2015 09:28:28 GMT

[quote user="Anthony Dennison"]After reading your article on shelter cats with murmurs David, I am wondering whether I should be investigating more with the RSPCA centre I do work for. Want some more data?[/quote]

Of course.

There's another in the pipeline about diagnosing Chlamydophila on conjunctival cytology too if you fancy poking some fluey cats in their eyes with a cotton bud...

RE: SNAP test grey zones

David Mills — Wed, 18 Nov 2015 09:23:43 GMT

[quote user="Martin Atkinson"]]I studied a series of 35 cats many of which many were sent for echo, the others didn't need to because they were in cardiac failure.[/quote]

Ah, was this the non-peer reviewed case series?

If you were in heart failure, how would you feel if the cardiologist said 'no need to send you for an echo, I can see you're in heart failure. We can do this blood test for £50 which will tell us what we already know, though?'

It is arguable that the only time it is necessary to echo (or investigate the heart) is when a cat is in heart failure.

[quote user="Martin Atkinson"]when did you last diagnose RCM in a cat period?[/quote]

5 cases this year and counting. If you don't look, you won't find.

[quote user="Martin Atkinson"]Sadly and you will say somewhat cynically, cardiologists don't like NT-proBNP because it puts them out of a job![/quote]

Ah yes, that must be it! Nothing to do with humbly realising is restrictions, false positives, false negatives and general use of the test...

RE: SNAP test grey zones

Martin Atkinson — Wed, 18 Nov 2015 09:15:45 GMT

[quote user="David Mills"]How do you know? You don't echo. And I assume you don't send all cardiac cats for echo.[/quote]I studied a series of 35 cats many of which many were sent for echo, the others didn't need to because they were in cardiac failure. There was a very close correlation in the vast majority with findings from echo, NT-proBNP and onset of symptoms of cardiac failure.

[quote user="David Mills"]When did you last diagnose RCM with proBNP?[/quote]when did you last diagnose RCM in a cat period?

Sadly I feel, and you will say somewhat cynically, cardiologists don't like NT-proBNP because it puts them out of a job! Despite the fact that I have 100% confidence in its predictive value in cats, I always offer echo if I get a high NT-proBNP but many owners cant afford it.

RE: SNAP test grey zones

Anthony Dennison — Wed, 18 Nov 2015 09:11:26 GMT

[quote user="David Mills"]It is nowhere near as sensitive as echo for prognosticating. When did you last diagnose RCM with proBNP?[/quote]

After reading your article on shelter cats with murmurs David, I am wondering whether I should be investigating more with the RSPCA centre I do work for. Want some more data?

[quote user="John Flynn"]The "pancreatic" lipase is a classic - look at the papers that compare its supposedly superior powers to "normal" lipase - that is when the latter is considered only as "positive" or "negative", while the former is afforded the luxury of 3 bands [/quote]

Alex German at Liverpool Vet School really dislikes pancreatic lipase tests. Calls them 'the best marketed test ever produced' because everyone uses them despite the fact they aren't that great.

[quote user="Robin Grimmer"]The canine parvo virus antigen one is pretty reliable, no?[/quote]

Can get false positives if they've been recently vaccinated.

RE: SNAP test grey zones

David Mills — Tue, 17 Nov 2015 22:46:54 GMT

[quote user="John Flynn"]actually evidence that vets hate numbers[/quote]

Point taken - maybe they love binary yes/no (or line/no line) more than numbers?? Or perhaps I've been too long in cardiology circles - we drown in measurements and numerical gradings, mostly largely superfluous.

Agree re tests. I'm always bemused why they don't provide the graph to interpret numerical results on like with PCVs.

RE: SNAP test grey zones

Robin Grimmer — Tue, 17 Nov 2015 22:12:01 GMT

[quote user="Andrew Kent"]

The only SNAP test I use is the FIV/FeLV.

The majority of the other snap tests are, at best, potentially misleading and at worst down right dangerous.

Andy

[/quote]

The canine parvo virus antigen one is pretty reliable, no?

RE: SNAP test grey zones

John Flynn — Tue, 17 Nov 2015 21:55:31 GMT

[quote user="David Mills"]

Any test is a trade off between sn/sp. Tests are calibrated to maximise one or the other, or give a reasonable compromise (rarely can any test give >80% of both) - they plot the results on a graph and read off what sn/sp they want.

Unfortunately, vets love numbers.

[/quote]

The reduction of quantitative data where the degree of elevation is proportional to the likelihood of disease to one or two "cut-off-points" means chucking away useful information. A numerical test result may well have a sensitivity or specificity much greater than 80% for a result at one or other extreme. Unfortunately published data consistently insists on introducing unhelpful dichotomous bands of "positive" and "negative". The "pancreatic" lipase is a classic - look at the papers that compare its supposedly superior powers to "normal" lipase - that is when the latter is considered only as "positive" or "negative", while the former is afforded the luxury of 3 bands

NT-proBNP papers are similarly frustrated with no access to the actual data to allow use of likelihood ratios even if you wanted.

Either I'm missing something (always very possible), or this insistence to only publish dichotomous, arbitrary "cut-off's", without access to the raw data, and then claim an overall "sensitivity" or "specificity" for a test is actually evidence that vets hate numbers.

RE: SNAP test grey zones

Francisco Gomez — Tue, 17 Nov 2015 21:33:37 GMT

[quote user="Andrew Kent"]

The only SNAP test I use is the FIV/FeLV.

The majority of the other snap tests are, at best, potentially misleading and at worst down right dangerous.

Andy

[/quote]

Did find a fiv+/leuk- in a SpeedDuo. Sent to external lab to find fiv-/leuk+.

RE: SNAP test grey zones

Andrew Kent — Tue, 17 Nov 2015 20:34:17 GMT

The only SNAP test I use is the FIV/FeLV.

The majority of the other snap tests are, at best, potentially misleading and at worst down right dangerous.

Andy

RE: SNAP test grey zones

David Mills — Tue, 17 Nov 2015 19:08:56 GMT

[quote user="Martin Atkinson"]I have found an almost perfect correlation between NT-proBNP and cardiac disease[/quote]

How do you know? You don't echo. And I assume you don't send all cardiac cats for echo. So that sentence is nonsense, gently be it put.

It is nowhere near as sensitive as echo for prognosticating. When did you last diagnose RCM with proBNP? What's the MST difference between that and boring old HCM?

RE: SNAP test grey zones

Martin Atkinson — Tue, 17 Nov 2015 19:01:19 GMT

[quote user="David Mills"]I have little faith in proBNP as it gives you far less information than echo (although it is better than nothing if you're not doing echo). [/quote] You haven't used it enough then. I have found an almost perfect correlation between NT-proBNP and cardiac disease and the results of echo especially with cats, less so with dogs. It is an important (and easy/inexpensive) prognostic indicator.

Was that patronising enough for you?

RE: SNAP test grey zones

David Mills — Tue, 17 Nov 2015 17:32:29 GMT

Unfortunately, vets love numbers. For something as complex as cardiac disease, I have little faith in proBNP as it gives you far less information than echo (although it is better than nothing if you're not doing echo).

RE: SNAP test grey zones

Thomas Johnson — Tue, 17 Nov 2015 15:37:54 GMT

[quote user="Martin Atkinson"]

I used to send PLi tests to the lab after a +ve SNAP but I have found little correlation of clinical symptoms with absolute values so I usually don't bother now.

[/quote]

I've never been persuaded of the benefit of the snap tests for pancreatic lipase, what am I going to differently in the 24hrs between when I'd have the result of the snap test, and when I'd have the result of the Spec cPL from the external laboratory? It will get fluids and pain relief as necessary, and a positive snap test wouldn't rule out a foreign body or other abdominal pathology. to quote from a recent CPD session I attended, "specific lipase is a great test for inflammation, just not necessarily of the pancreas."

RE: SNAP test grey zones

Martin Atkinson — Tue, 17 Nov 2015 10:59:07 GMT

[quote user="Anthony Dennison"] Martin spoke about the proBNP snap test having a large grey zone - the same could be said about the snap fPLI/cPLI tests. How many people use snap tests 1st choice over sending bloods away for a specific value? [/quote]

I think there is a fundamental difference here that a workbench test gives you an immediate answer so you can act upon it as there is no time to wait for a test to come back from a lab. With cardiac disease it is pretty obvious if an animal is in acute cardiac failure so you don't need an immediate answer. I believe that fPLi is not a reliable indicator of pancreatic disease in cats even if it is sent to the lab. A positive test is however a positive with PLi. The biggest issue is interpreting the colour change as it is not always clear. If however I'm in doubt then I consider it positive.

I used to send PLi tests to the lab after a +ve SNAP but I have found little correlation of clinical symptoms with absolute values so I usually don't bother now.