ACVIM MMVD Stage Interpreter (A, B1, B2, C, D)

ACVIM Stage for Myxomatous Mitral Valve Disease, and the Four Numbers That Separate B1 From B2

The 2019 ACVIM consensus stages myxomatous mitral valve disease as A, B1, B2, C or D. Only one boundary is arithmetic: the one between B1 and B2, which decides whether a dog with no clinical signs starts pimobendan. The consensus sets it with four criteria at once — murmur 3/6 or louder, LA:Ao 1.6 or more, LVIDDN 1.7 or more, breed-adjusted VHS above 10.5 — and three of the four are the entry criteria of the EPIC trial that produced the recommendation. This page applies them, shows which are met, and gives the radiographic route for a practice without an echocardiogram.

Staging systems are classification tools built on populations, and this page applies the published criteria to the numbers you enter. A stage is not a diagnosis, it does not substitute for imaging or a full clinical picture, and the guideline it follows is named on the page so you can check the version you are working to.

Stage the dog, and see which B2 criteria are reached

ACVIM 2019 consensus stages A, B1, B2, C and D
Decides the stage on its own for A, C and D. Stage C is defined by current or past signs, so a dog treated successfully last year is still stage C; stage D is signs “refractory to standard treatment for Stage C”, a treatment-response definition rather than a measurement.
The B2 criterion is 3/6 or louder. The EPIC abstract records no murmur criterion at all, so this is the one B2 criterion that did not come from the trial the recommendation rests on. Murmur intensity tracks regurgitant severity only loosely.
Measured in the right-sided short-axis view in early diastole — the consensus specifies view and timing because other conventions give different numbers. Threshold 1.6 or more. With no echocardiogram, leave this below 1.6 for the radiographic route.
Left ventricular internal diameter in diastole divided by body weight in kilograms to the power 0.294 — the allometric normalisation that makes one threshold usable across a 2 kg and a 40 kg dog. Threshold 1.7 or more.
A breed-adjusted VHS above 10.5 is one criterion; a general breed VHS of 11.5 or more is the separate no-echo substitution. Two thresholds, two purposes, routinely confused. Measure it against the right breed reference.
The four B2 criteria as four gauges, each with its consensus threshold ticked and the shaded bar appearing from the threshold rightwards only when that criterion is met; three of the four thresholds are also EPIC's entry criteria. The five boxes below are the stages, with an arrow on the one these answers select. An arrow between B1 and B2 means one echocardiographic criterion is reached and the other is not.
ACVIM stage B2. Asymptomatic MMVD with cardiac remodelling, and all four consensus criteria are reached.Example

an 8-year-old Cavalier King Charles Spaniel with no clinical signs, a grade 3/6 left apical systolic murmur, LA:Ao 1.7, LVIDDN 1.8 and a breed-adjusted VHS of 11.0

Advertisement

One boundary is arithmetic; the other three are clinical

Stage B2 needs all of: murmur ≥ 3/6 · LA:Ao ≥ 1.6 · LVIDDN ≥ 1.7 · breed-adjusted VHS > 10.5  |  no echo: general breed VHS ≥ 11.5 or VLAS ≥ 3
A, C and D
Clinical definitions, not measurements. A is a predisposed breed with no apparent structural abnormality. C is current or past signs of heart failure from MMVD, so it does not reverse. D is signs refractory to standard stage C treatment. None of the four numbers moves a dog between them, and B1 against B2 is the only arithmetic boundary in the scheme.
1.6, 1.7 and 3/6
LA:Ao at or above 1.6 in the right-sided short axis in early diastole, and LVIDDN at or above 1.7 — the left ventricular internal diameter in diastole divided by body weight to the power 0.294, the allometric normalisation that lets one threshold serve a 2 kg and a 40 kg dog. Both are EPIC entry criteria; the 3/6 murmur is not, since the trial abstract records no murmur criterion at all.
10.5 and 11.5
Two different thresholds, routinely confused. Breed-adjusted VHS above 10.5 is one of the four criteria used alongside echocardiography; a general breed VHS of 11.5 or more is the substitute allowed where echocardiography is unavailable, set higher because a radiograph alone must be more specific. VLAS of 3 or more is accepted by the same route.
pimobendan
0.25 to 0.3 mg/kg PO q12h in stage B2, Class I on strong evidence, from EPIC: median 1228 against 766 days to heart failure, cardiac death or euthanasia, hazard ratio 0.64 (0.47 to 0.87), P = .0038. This page is arithmetic on a published scheme and prescribes nothing.

Worked example

an 8-year-old Cavalier King Charles Spaniel with no clinical signs, a grade 3/6 left apical systolic murmur, LA:Ao 1.7, LVIDDN 1.8 and a breed-adjusted VHS of 11.0
SETTLE THE CLINICAL STAGE FIRST. No current or past signs of heart failure, so not C and not D. There is a murmur, so not A. That leaves B, and only then do the four numbers matter.
MURMUR AND THE TWO ECHO CRITERIA. 3/6 against 3/6 or louder: met, exactly on the boundary. LA:Ao 1.7 against 1.6 and LVIDDN 1.8 against 1.7: each met by 0.1.
VHS, AND AGAINST WHICH BREED. 11.0 against a breed-adjusted criterion above 10.5: met by 0.5. But Lamb et al. put the Cavalier mean at 10.6 plus or minus 0.5 and Bagardi et al. at 10.08 plus or minus 0.56, so this dog is 0.8 standard deviations above its breed mean on one reference and 1.6 on the other. Both agree here; in a borderline dog they would not.
ALL FOUR MET, SO B2. The consensus asks for all of them “ideally” before treatment is started, and recommends pimobendan 0.25 to 0.3 mg/kg q12h. Three of the four are within 0.5 of their thresholds, which is worth saying aloud: this dog crossed the line rather than sailing past it, and repeating the echocardiogram in three months would tell you which direction it is going.
Advertisement

The B2 criteria and where each one came from

CriterionThresholdAn EPIC entry criterion?
Murmur intensity3/6 or louderNo – the EPIC abstract records no murmur criterion
LA:Ao, right short axis, early diastole1.6 or moreYes
LVIDDN1.7 or moreYes
Breed-adjusted VHSabove 10.5Yes
General breed VHS, no echocardiography11.5 or moreNo – a consensus substitution
VLAS, no echocardiography3 or moreNo – a consensus substitution
Thresholds from Keene et al. 2019; the EPIC column from the Boswood et al. 2016 abstract, which lists LA:Ao at least 1.6, LVIDDN at least 1.7 and VHS above 10.5 and no murmur grade.

What EPIC actually showed

PimobendanPlacebo
Dogs360 randomised in total
Median days to the primary endpoint1228 (95% CI 856 to not reached)766 (667 to 875)
Hazard ratio0.64 (95% CI 0.47 to 0.87)reference
P value0.0038
Primary endpointonset of congestive heart failure, cardiac death or euthanasia
From the EPIC abstract, read 18 November 2026; the full paper was not reachable. The difference in medians is about 15 months, which is the figure to quote to an owner rather than the hazard ratio.

Why the B1 to B2 line is drawn with four numbers instead of one

Most of the ACVIM scheme is clinical: a dog is stage A because of its breed, stage C because it has had heart failure, stage D because treatment stopped working. One boundary is different, and it is the one this page exists for: the line between B1 and B2 decides whether a dog with no symptoms starts a drug it will then take for life.

The consensus draws that line with four criteria at once because of the trial underneath it. EPIC enrolled 360 dogs on three measurements — LA:Ao at least 1.6, LVIDDN at least 1.7, VHS above 10.5 — and found a median of 1228 days to heart failure, cardiac death or euthanasia on pimobendan against 766 on placebo. That is roughly fifteen months of extra preclinical time, in dogs defined that way. The instruction that ideally all these criteria should be met is a statement about who the evidence applies to: a dog meeting one criterion is not a dog in the trial.

The murmur criterion is the odd one out, and saying so is more useful than hiding it. The consensus asks for 3/6 or louder; the EPIC abstract records no murmur criterion whatsoever, and murmur intensity is a loose proxy for regurgitant volume that can fall as disease advances. So a dog with both echocardiographic criteria met and a quieter murmur matches the trial population even though it fails the consensus list. This page reports that as a B2 heart with a quiet murmur and says which criterion it set aside.

The radiographic route matters more than it is given credit for, because most dogs with MMVD are never seen by anyone with an echocardiogram. A general breed VHS of 11.5 or more, or a VLAS of 3 or more, substitutes; both sit higher than the echocardiographic criteria because a single radiographic number has to be more specific when nothing else is checking it. The caveats: VHS varies by breed by more than a vertebra between the Yorkshire Terrier and the Bulldog, two competent readers of one film can differ by half a vertebra, and VLAS beat VHS for detecting atrial enlargement in Duler et al. 2021 (90.3% against 75.8% sensitivity) but not significantly for stage B2 (0.84 against 0.78, P = .17).

Two things the staging does not do. It says nothing about pulmonary hypertension or atrial fibrillation, which arrive on their own schedule. And it says nothing about the kidney, which every drug used from stage C onwards acts on or is cleared by: a dog on furosemide, pimobendan and an ACE inhibitor needs its renal stage known, dose adjustment where the stage demands it, and a view on volume that a standard maintenance fluid rate will not give. Human left ventricular mass indexing uses body surface area rather than the allometric exponent that makes LVIDDN work across a fortyfold weight range.

Frequently asked questions

What are the ACVIM criteria for stage B2?

Four, and the consensus asks for all of them ideally: a murmur of 3/6 or louder, an LA:Ao of 1.6 or more in the right-sided short axis in early diastole, an LVIDDN of 1.7 or more, and a breed-adjusted VHS above 10.5. Without echocardiography, a general breed VHS of 11.5 or more or a VLAS of 3 or more may substitute.

Is there a newer ACVIM consensus than 2019?

Not as of 18 November 2026. Keene et al. 2019 (J Vet Intern Med 33:1127-1140) is still the current statement, and a Veterinary Times review dated April 2025 calls it the most recent ACVIM consensus statement while arguing that newer work has made parts of it slightly obsolete. The work it means includes DELAY, VALVE, BESST and the MINE score, none of which changes the B1 to B2 criteria.

When does pimobendan start in MMVD?

The consensus recommends it in stage B2 at 0.25 to 0.3 mg/kg PO q12h, Class I on strong evidence from the EPIC trial. It is not recommended in B1, where no published trial has shown a benefit. This page is arithmetic on a published scheme; the dose belongs to the attending clinician.

My dog meets three criteria but not the fourth. Is it B2?

It depends which one. Both echo criteria met with a quieter murmur still matches the EPIC population, because EPIC had no murmur criterion. Both met with only the VHS short favours the echocardiogram. One echo criterion met and the other not does not satisfy the definition, and a repeat in three to six months beats a decision.

Can I stage B1 against B2 on a radiograph alone?

Yes, by the route the consensus provides: a general breed VHS of 11.5 or more, or a VLAS of 3 or more. Know its limits — a VHS cut-off of 12 reached only 37% sensitivity at 99% specificity in the literature Baisan and Vulpe collect, so it misses dogs rather than over-calling them. Note also that stage C does not reverse: the consensus defines it by current or past signs, so a dog whose oedema resolved two years ago is still stage C.

Related calculators

References

  1. Keene BW, Atkins CE, Bonagura JD, Fox PR, Häggström J, Luis Fuentes V, Oyama MA, Rush JE, Stepien R, Uechi M. ACVIM consensus guidelines for the diagnosis and treatment of myxomatous mitral valve disease in dogs. J Vet Intern Med 2019;33(3):1127–1140. Read 18 November 2026. Stage B2 criteria, verbatim: murmur “≥3/6”, LA:Ao right short axis in early diastole “≥1.6”, LVIDDN “≥1.7”, breed-adjusted VHS “>10.5”, and “ideally, all these criteria should be met before initiating treatment”. Without echocardiography, a general breed VHS of ≥11.5 may substitute and “VLAS values of ≥3 likely identify Stage B2 MMVD”. Pimobendan in B2 0.25–0.3 mg/kg PO q12h, Class I. The panel warns of “breed differences in normal vertebral heart scales” and gives no breed table. Still the current statement; no later ACVIM revision was found on 18 November 2026.
  2. Boswood A, Häggström J, Gordon SG, et al. Effect of Pimobendan in Dogs with Preclinical Myxomatous Mitral Valve Disease and Cardiomegaly: The EPIC Study. J Vet Intern Med 2016;30(6):1765–1779 (PMID 27678080). Abstract read 18 November 2026; the full paper was not reachable. 360 dogs, enrolled on LA:Ao at least 1.6, LVIDDN at least 1.7 and VHS above 10.5 — three of the four ACVIM B2 criteria. Median time to heart failure, cardiac death or euthanasia 1228 days on pimobendan against 766 on placebo, hazard ratio 0.64 (95% CI 0.47 to 0.87), P = .0038. The abstract records no murmur criterion, so the 3/6 murmur in the consensus did not come from this trial.
  3. Diagnosis and management of mitral valve disease in dogs, Veterinary Times, dated 1 April 2025, read 18 November 2026 as a secondary summary. Confirms Keene et al. 2019 as “the most recent ACVIM consensus statement” while arguing newer work has made parts of it “slightly obsolete”, and reproduces the four B2 criteria unchanged. Names the trials since: DELAY (2020), spironolactone with benazepril did not significantly delay heart failure onset but improved remodelling; VALVE (2020), no survival benefit from adding an ACE inhibitor to furosemide and pimobendan; a 2022 meta-analysis finding no significant reduction in heart failure risk from ACE inhibitors in preclinical dogs with cardiomegaly; BESST (2021); and the MINE score (2021, updated at ECVIM 2023). Warns that LA:Ao can mislead through breed and inter-operator variability.
  4. Baisan RA, Vulpe V. Vertebral Heart Score and Vertebral Left Atrial Size as Radiographic Measurements for Cardiac Size in Dogs: A Literature Review. Animals 2025;15(5):683, doi 10.3390/ani15050683. Read 18 November 2026, and the source of every breed figure and most VLAS figures on these pages, each attributed there to a named primary study which was not read individually. Means ± SD in vertebrae, right lateral: general dog 9.7 ± 0.5; Boxer 11.6 ± 0.8, Labrador 10.8 ± 0.6, Cavalier 10.6 ± 0.5, Doberman 10.0 ± 0.6, German Shepherd 9.7 ± 0.7, Yorkshire Terrier 9.7 ± 0.5 (Lamb et al. 2001); Labrador 10.39 ± 0.15 (Bodh 2016); Cavalier 10.08 ± 0.56 (Bagardi 2022); Beagle 10.5 ± 0.4; Whippet 11.0 ± 0.5; Greyhound 10.5 ± 0.1; Pomeranian 10.5 ± 0.9; Boston Terrier 11.7 ± 1.4; Bulldog 12.7 ± 1.7; Shih Tzu 9.5 ± 0.6. Disagreeing cut-offs for cardiomegaly in MMVD: 12 vertebrae 37% sensitivity and 99% specificity, 10.8 gives 91.1% and 69.5%, 11 in Cavaliers 80% and 74%. Concludes that VHS alone should be read with care and cardiomegaly confirmed on echocardiography.

CalcEngines veterinary calculators are arithmetic tools for qualified veterinary professionals and for informed animal owners. They are not a prescribing decision, not a diagnosis, and not a substitute for examining the animal. Dose ranges are quoted from named published sources and those sources disagree with each other more often in veterinary medicine than in human medicine; where they do, the page shows the range. Species differences are not details — a dose that is routine in a dog can be fatal in a cat or a rabbit. Always confirm against the product literature, your own formulary and your practice protocol, and if an animal may have been poisoned contact a veterinary poison service or your vet immediately.