scott@vtx-cpd.com

Hi Ruth,

Thanks for sharing this case. I can understand your frustration, particularly as you’ve done a really thorough investigation, established a diagnosis with histopathology and bile culture, and demonstrated a good response to antimicrobial treatment, only to see the liver enzymes increase again once antibiotics are discontinued.

I would also be interested to know a little more about the patient. What is the signalment, particularly the breed, age and sex? Have you investigated leptospirosis, and were any concurrent gastrointestinal or pancreatic abnormalities identified? I would also be interested in whether the original bile cytology identified intracellular bacteria or whether the organisms were predominantly extracellular, and whether the gall bladder sediment has changed on subsequent ultrasound examinations.

I think the first thing to say is that this is a fairly convincing case of bacterial neutrophilic cholangitis or cholangiohepatitis. The acute presentation, particularly the fact that the dog was clinically unwell, together with the marked increases in ALT and bilirubin, certainly suggests an infectious component. The histopathology showing predominantly neutrophilic periportal inflammation, together with the isolation of E. coli from bile, makes an ascending bacterial infection very likely. The repeated biochemical improvement with antibiotics also supports this, although we have to be careful about assuming that every subsequent increase in ALT necessarily represents recurrence of infection.

I think all the treatment you’ve provided so far has been entirely appropriate, including the initial use of Synulox and subsequently the introduction of a fluoroquinolone. The initial culture demonstrated susceptibility to both drugs, and the improvement in liver enzymes following treatment was encouraging. I certainly wouldn’t look at this case and think that you should have managed the initial presentation differently.

One thing that interests me is the speed with which the liver enzymes have increased following discontinuation of antibiotics. After the initial 10 day course of Synulox, the ALT increased quite dramatically, and something similar happened after the six week course. That makes me wonder whether the infection was ever completely eliminated or whether there is an underlying biliary abnormality allowing infection to persist or recur. However, I think there is another possibility here that is equally important to consider.

The biopsy findings are particularly interesting because, although the neutrophilic component supports an acute infectious or inflammatory process, the presence of periportal fibrosis suggests that there may also be a more chronic component to the disease. We sometimes see patients that initially present with quite dramatic acute hepatobiliary disease, but subsequently develop persistent inflammation even after the original insult has resolved. That is probably what I would be most suspicious of in this case, particularly given the repeated biochemical response to treatment followed by further increases in ALT.

One of the main points I always try to emphasise with acute liver disease is that an acute insult does not necessarily mean the problem will remain entirely acute. There is always the possibility of ongoing inflammation and, in some cases, progression towards chronic hepatobiliary disease. The original bacterial infection may have been the initiating event, but it does not necessarily follow that bacteria are responsible for every subsequent increase in liver enzymes. We could now be dealing with an ongoing inflammatory response that is no longer directly associated with active infection.

In terms of antimicrobial duration, yes, some of these cases do require prolonged treatment. For bacterial neutrophilic cholangitis, I generally recommend a minimum of six to eight weeks of culture directed antimicrobial therapy, and occasionally longer depending on the clinical and biochemical response. Courses extending to two or even three months are certainly not unheard of, particularly in complicated cases where infection is difficult to eradicate. However, I would be reluctant to simply prescribe another three months of marbofloxacin without first considering why the abnormalities keep returning. Although the E. coli was reportedly susceptible to both amoxicillin clavulanate and marbofloxacin, susceptibility in vitro does not necessarily guarantee eradication of an infection within the biliary system, particularly if there is impaired bile drainage, inspissated bile, or an underlying structural abnormality.

I would be interested in reviewing the original ultrasound findings in more detail, particularly the appearance of the gall bladder wall, common bile duct and intrahepatic biliary tree. You mentioned some inspissated bile and a small amount of free fluid adjacent to the liver. Neither finding necessarily establishes significant biliary disease, but in a patient with recurrent bacterial cholangitis I would want to reassess these structures carefully. I would also be interested in whether there was any evidence of gall bladder mucosal hyperplasia, cholelithiasis, biliary dilatation or changes involving the pancreas or proximal duodenum that might contribute to ascending infection.

The other thing I would seriously consider is repeating the cholecystocentesis rather than continuing to alternate between Synulox and marbofloxacin. I appreciate that this may seem a little concerning given the previous infection and the possibility of introducing further complications, but I think it could be very valuable here. We know E. coli was present initially, but we do not know whether it is still present, whether a different organism has subsequently become involved, or whether the biliary infection has actually resolved and we are now dealing with residual or recurrent sterile inflammation.

If you are contemplating another prolonged course of antibiotics, I would much rather establish whether bacteria are still present. Repeating bile cytology and aerobic and anaerobic culture would be a relatively less invasive way of investigating this compared with repeating liver biopsies, assuming the gall bladder is suitable for safe sampling. It would also allow you to confirm that you are still using an appropriate antibiotic rather than assuming the original susceptibility pattern remains relevant. Ideally, I would obtain those samples before restarting antimicrobials, provided the patient remains clinically stable.

If the repeat bile culture is positive, then I think there is a reasonable argument for another prolonged course of appropriately selected antibiotics, potentially extending to eight to twelve weeks depending on the findings and response. I would also want to investigate whether there is an underlying reason why the infection is persisting. If the culture is negative, particularly if the cytology is also unremarkable, then I would become increasingly suspicious that we are dealing with a more chronic inflammatory hepatobiliary process rather than persistent bacterial infection. Of course, a negative culture does not completely exclude infection, particularly following recent antimicrobial treatment, but it would certainly influence my thinking.

Regarding the most recent results, an ALT of 128 is a very different situation from an ALT of 1,349, particularly if the dog is clinically well and the bilirubin has normalised. Assuming 128 represents only a mild elevation relative to your laboratory’s reference interval, I would not automatically restart marbofloxacin at this stage. I would be inclined to repeat the biochemistry in approximately one to two weeks, ideally including ALT, ALP, GGT and bilirubin, to establish whether this is genuinely increasing again or simply a small residual fluctuation following treatment. I think there is a reasonable chance that we may be overinterpreting a relatively minor elevation given the severity of the previous abnormalities.

If the ALT is clearly increasing again, particularly if we start seeing accompanying increases in bilirubin or cholestatic enzymes, then I would be much more concerned about ongoing hepatobiliary disease. At that point, I would favour repeat abdominal ultrasound and, if appropriate, repeat bile sampling for cytology and aerobic and anaerobic culture before committing to another prolonged antimicrobial course.

I would also continue the Denamarin and Destolit for now. I am comfortable using these together, provided there is no evidence of complete extrahepatic biliary obstruction. Ursodeoxycholic acid may be particularly useful here because of its choleretic properties, although obviously it is not going to eliminate an established bacterial infection on its own.

The other consideration, and perhaps the most important one for me, is whether we now need to think about treating an underlying chronic inflammatory component. The periportal fibrosis on histopathology suggests chronicity, and although the neutrophilic inflammation is consistent with an ascending bacterial process, it does not necessarily exclude an ongoing inflammatory response after the infection has resolved.

If we can reasonably exclude persistent infection, then I think there may be a role for introducing anti-inflammatory corticosteroid therapy rather than continuing antibiotics indefinitely. I would not necessarily jump straight to prednisolone at this stage, because I would be uncomfortable introducing immunosuppression while there is still a realistic possibility of active bacterial infection. However, if repeat bile culture is negative and the liver enzymes continue to increase, I think that becomes a reasonable discussion.

We do occasionally encounter patients where the original insult appears to have been infectious, but the subsequent clinical problem is ongoing inflammatory hepatobiliary disease. In those circumstances, continuing to prescribe antibiotics may not address the underlying pathology. A carefully monitored trial of anti-inflammatory treatment could potentially be more useful, although I would want to be reasonably confident about the diagnosis before taking that approach.

You could also consider repeating the liver biopsy, particularly if the abnormalities become more significant or the diagnosis remains uncertain. That would allow you to determine whether the original neutrophilic inflammation has resolved, whether there is now a more chronic lymphoplasmacytic or mixed inflammatory infiltrate, and whether the fibrosis has progressed. It would also help establish whether corticosteroid treatment is justified. Obviously, repeating a biopsy is more invasive, and I appreciate there may be financial considerations, so I would probably favour repeating cholecystocentesis first if the patient is otherwise doing well.

One other question I have is whether you have tested for leptospirosis. I appreciate that the isolation of E. coli from bile and the histopathology make ascending bacterial cholangitis the leading diagnosis, but given the severity of the original hepatocellular injury and hyperbilirubinaemia, I would still be interested in whether leptospirosis was investigated or excluded. It would not be my leading differential at this stage, but I think it is worth considering in a dog with such a dramatic initial hepatic presentation.

I would also be interested to know whether the original bile cytology identified intracellular bacteria or whether the organisms were predominantly extracellular. You mentioned that numerous bacteria were present without significant active inflammation, which is interesting, particularly given the neutrophilic inflammation identified on liver histopathology. I would also be interested in the dog’s breed and age, whether any concurrent gastrointestinal or pancreatic abnormalities were identified, and whether the gall bladder sediment has changed on subsequent ultrasound examinations.

So, if this were my patient, and assuming she is currently clinically well, I would hold off restarting marbofloxacin, repeat the liver enzymes in one to two weeks, and see what direction things are moving. If they remain stable or normalise, I would continue monitoring. If they increase substantially again, I would investigate for persistent biliary infection rather than automatically prescribing another three months of antibiotics.

If repeat cholecystocentesis confirms persistent infection, then a further prolonged course of culture directed antibiotics would be appropriate. If it does not, then I think we need to seriously consider whether we are dealing with a chronic inflammatory process that may ultimately require anti-inflammatory treatment rather than continued antimicrobial therapy.

I certainly wouldn’t regard this as a treatment failure on your part. You’ve established the diagnosis and obtained a very good response to treatment. The challenge now is determining whether we are dealing with an incompletely resolved infection, repeated ascending infection, or residual hepatobiliary inflammation. Those are slightly different problems, and I think distinguishing between them is going to be much more useful than simply extending the antibiotic course indefinitely.

Let me know how the next set of bloods looks. I’d be interested to hear how this one progresses, particularly whether you have investigated leptospirosis and whether repeating the bile culture might be an option.

Scott 🙂

Replying to Kirsten Hearn 09/10/2026 - 16:58

Hi Kirsten,

Just a further update on this. I’ve also heard back from Janette, another member of our anaesthesia team, who has kindly put together a really useful seven-page document covering the preparation and administration of dopamine, dobutamine, norepinephrine and ephedrine.

It includes practical dilution tables based on patient weight, suggested dose ranges, worked calculations and guidance on choosing the appropriate drug depending on the underlying cause of hypotension. She has also included some important safety considerations around preparing and administering CRIs.

One of the approaches she outlines is adjusting the concentration according to patient weight so that the number displayed on the syringe driver in mL/hr corresponds directly to the dose in micrograms/kg/min, which is very much what you were originally hoping to achieve. Her tables currently use 100 mL preparations, so these would need adapting for your 50 mL syringes.

I think you’ll find it really useful alongside Kerry’s suggestions, particularly when developing your own protocol.

If you can let me know the best email address to reach you on, I’ll happily send you a copy of Janette’s document.

Scott x

Hi Kirsten,

I spoke to one of our ECC team about this. Her feeling was that, particularly for a team that is relatively new to vasoactive CRIs, dopamine is a very sensible place to start. It has a rapid effect, is easy to titrate, is inexpensive and is generally readily available.

She suggested standardising your dopamine CRI to a final concentration of 1 mg/mL. You can make that up in your 50 mL syringe driver; the actual volume of stock dopamine you add will obviously depend on the concentration of the product you have.

For routine hypotension under GA, she keeps the dopamine range quite simple at 5–7.5 micrograms/kg/min, starting at 5 and titrating according to the blood pressure response.

At a final concentration of 1 mg/mL, the calculation is:

mL/hr = bodyweight (kg) × dose (micrograms/kg/min) × 0.06

So you could make a simple laminated chart beside the syringe driver:

The important bit before reaching for the CRI is still to assess why the patient is hypotensive: anaesthetic depth, volume status, heart rate/rhythm, ventilation and underlying disease, rather than treating the blood pressure number in isolation.

She tends to reserve dobutamine for situations where poor contractility/cardiac dysfunction is specifically suspected, rather than using it routinely for uncomplicated GA hypotension. If you can get norepinephrine/noradrenaline, she felt it was definitely worth having available, particularly when vasodilation/vasoplegia is the predominant problem, such as in septic patients. She also uses ephedrine, although was a little more cautious about introducing that initially to teams with less experience of these drugs.

I think your idea of standardising this is excellent. Having one standard dopamine concentration and a laminated weight/rate chart beside the driver removes most of the maths at exactly the point when people don’t want to be doing calculations!

I hope that helps.

Scott 🙂

Hi Raquel,

Great question. Yes, I have used both Spec fPL and the newer Catalyst Pancreatic Lipase assay, and I think the quantitative in-house test is a useful development. I would be quite happy using the Catalyst PL in the way you describe, alongside the history, clinical findings and ultrasound—rather than routinely sending every sample to the reference laboratory for Spec fPL.

One important distinction is that the Catalyst pancreatic lipase test isn’t the same thing as a conventional serum lipase activity assay. IDEXX designed it specifically to measure pancreatic lipase and to align clinically with their Spec fPL/Spec cPL assays. In their validation data, the feline Catalyst PL showed 87.5% overall categorical agreement with Spec fPL, with good analytical precision. So if you already have the Catalyst in practice, getting a quantitative result in less than 10 minutes is genuinely useful and, as you say, may also be considerably cheaper than sending Spec fPL away.

The change in the feline cut-off is also interesting, but there is a bit of history behind it. The older literature commonly used ≥5.4 µg/L as the Spec fPL threshold for pancreatitis. One clinical study using that threshold reported sensitivity of 79.4% and specificity of 79.7% when cats considered definitely/probably pancreatitis were compared with those considered probably/definitely not pancreatitis.

IDEXX subsequently changed its interpretation following a re-evaluation of the Spec fPL assay. That analytical validation established an upper reference limit of 4.4 µg/L in healthy cats and calculated 8.8 µg/L as a diagnostic decision threshold designed to give approximately 99% specificity against the healthy population. IDEXX therefore now essentially divides results into three groups: ≤4.4 µg/L = pancreatitis unlikely/within the reference range; 4.5–8.7/8.8 µg/L = equivocal, pancreatitis possible; and around ≥8.8/8.9 µg/L = consistent with pancreatitis, depending on whether you are looking at the Catalyst or reference-laboratory reporting convention.

I actually prefer thinking about the test that way rather than treating pancreatic lipase as a binary positive/negative result.

The important caveat is that increasing the diagnostic threshold doesn’t somehow make pancreatic lipase a definitive test for pancreatitis. What it primarily does is make a markedly increased result more specific. There is inevitably a trade-off between sensitivity and specificity as we move the cut-off. A cat with a PL of 5 or 6 doesn’t suddenly cease to have pancreatitis because it falls below 8.8; it means that the biochemical evidence alone is insufficiently specific for us to confidently attribute the clinical signs to pancreatitis.

That is particularly important in cats because pancreatic abnormalities can coexist with gastrointestinal, hepatobiliary and other disease. IDEXX themselves make this point in their current diagnostic algorithm: if another disease is identified, pancreatic involvement may be concurrent rather than necessarily being the primary cause of the clinical presentation.

So I approach these results roughly as follows. If I have a cat with convincing clinical signs, compatible ultrasonographic pancreatic changes and a PL ≥8.8, I am considerably more confident that pancreatitis is genuinely part of the clinical problem. If the result is 4.5–8.7, I regard it as supporting information but certainly wouldn’t diagnose pancreatitis from that result alone. I look particularly carefully at the ultrasound, clinical presentation and concurrent disease. If it is ≤4.4, clinically important pancreatitis becomes less likely, but I still wouldn’t allow one laboratory result to completely overrule a very convincing clinical picture.

That is also why I like what you are already doing by combining pancreatic lipase with ultrasound and the clinical presentation. Neither test is perfect. Ultrasound is operator-dependent and pancreatic changes are not synonymous with clinically significant pancreatitis, while pancreatic lipase gives us biochemical evidence of pancreatic injury but doesn’t tell us whether the pancreas explains everything that is happening to that particular cat. Even studies comparing pancreatic lipase with histopathology have demonstrated the limitations of trying to make pancreatitis a simple positive/negative diagnosis.

The other reason I prefer the quantitative Catalyst test to the SNAP when it is readily available is simply that I get much more information from a number. The SNAP fPL was deliberately developed primarily as a screening/rule-out test and prioritises sensitivity; IDEXX did not change the SNAP when they revised the quantitative Spec fPL reference intervals. A value of 3, 6 and 15 µg/L gives me substantially more clinical information than simply knowing whether a SNAP is normal or abnormal. It is also much more useful if I subsequently want to look at the direction of travel in a patient.

Would I still send a Spec fPL to the external laboratory? Occasionally, yes. If I had an unusual or difficult case where the Catalyst result didn’t make sense in the context of everything else, or I particularly wanted laboratory confirmation, I might. But I don’t think every elevated Catalyst PL needs to be followed by Spec fPL. That rather defeats the purpose of having a validated quantitative patient-side assay, particularly given the good agreement between the two methods. I would spend the money on the rest of the diagnostic investigation where appropriate.

One small technical point worth remembering is that Catalyst PL and Spec fPL don’t report identical units—Catalyst PL is reported in U/L whereas Spec fPL is µg/L—so although the feline numerical decision limits were deliberately aligned, I wouldn’t describe the assays as analytically interchangeable measurements of exactly the same thing. They are different assays intended to give clinically concordant interpretation.

So overall, yes, I think the in-house quantitative PL is a useful test and I am very comfortable using it. I actually prefer the newer three-zone interpretation to the old idea that anything above 5.4 automatically means pancreatitis. It forces us to retain an equivocal zone and gives greater weight to markedly increased results, while still making us interpret the number in the context of the cat in front of us.

And I think your current approach—history + clinical signs + quantitative pancreatic lipase + good-quality ultrasound—is exactly the sensible way to use it. I would never diagnose or exclude feline pancreatitis from any one of those pieces of information in isolation.

Hope that helps!

Scott 🙂

Hi Silvana,

I did a bit more digging into this as it really interested me. It looks as though GingiPro is available through Bova in the UK as well as Australia, although there are some important considerations around its use.

The combination contains EIDD-1931, doxycycline and meloxicam, which raises a few questions for me. Ideally, before considering something like this, I think I would want to be confident that concurrent periodontal disease had been properly assessed and treated, including dental radiographs and extraction of diseased teeth where appropriate. I would also want confirmation of FCV by PCR rather than using an antiviral empirically, and obviously the individual cat needs to be an appropriate candidate for an NSAID.

There are also some wider stewardship questions around whether doxycycline is necessary in every case and around antiviral use in cats without confirmed calicivirus. EIDD-1931 itself is interesting, but potential adverse effects and owner exposure/handling need consideration, particularly as there are teratogenicity concerns. Applying a medication directly to the gingiva of a cat with a very painful mouth may not always be particularly straightforward either!

The other limitation at the moment seems to be the evidence base. There are apparently good anecdotal results, but the optimal duration of treatment has not yet been properly studied or published. Molnupiravir/EIDD-related antiviral treatment on its own has also been investigated in FCV-positive cats with encouraging results, so I suspect this is an area we are going to hear considerably more about.

So, potentially very interesting, and I can certainly see a role for this type of antiviral treatment in carefully selected FCV-positive cats, but I think I would currently be a little cautious about the one-size-fits-all combination product approach until we have more published data.

Thanks for bringing it up — I genuinely hadn’t come across it before!

Scott