Differentiating Cardiac and Noncardiac Causes of Nonhemorrhagic Ascites with NT-proBNP, cTnI and POCUS in dogs | VETgirl Veterinary Continuing Education Podcasts

May 2024

Dr. Amy Kaplan, DACVECC here today! In today’s VETgirl online veterinary CE podcast, we’re going to explore the various diagnostics tools available to us to workup nonhemorrhagic ascites in dogs beyond our initial physical examination. Now in the ER, when I tap an abdomen and it comes back as blood, I’m all over it – the majority of the time, it’s often secondary to hemangiosarcoma or less common causes such as hemangiomas or hematomas (which DO happen!). Less commonly, I’ll see patients present with abdominal effusion that is considered a nonhemorrhagic ascites (what we’ll call NHA from now on) – this is fluid with a packed cell volume of < 10%). A variety of diseases can cause NHA including right-sided congestive heart failure, neoplasia, liver failure, hypoalbuminemia, pancreatitis, chyloabdomen, uroabdomen, or septic peritonitis. So, when it’s NOT blood, how do you figure out the cause of the ascites?

Differentiating cardiac from non-cardiac causes of nonhemorrhagic ascites (or NHA) can be challenging. Physical examination may not always show us strong indicators to suggest the most likely cause of the ascites. If we think about cardiac causes of ascites, well, echocardiography is the definitive method for detecting and quantifying heart disease in dogs but it can’t always be included in the initial workup for NHA for a variety of reasons, including cost and availability. (Often times other diagnostics such as complete blood count, chemistry, urinalysis, chest radiographs, and abdominal ultrasound are performed first, and these costs can really add up!).

So, the question for today is, would assessment of cardiac biomarkers and point-of-care ultrasound (POCUS) help provide additional insight into the cause of nonhemorrhagic ascites (again NHA), as these options may be more readily available in some areas and cost less? Specifically, N-terminal probrain natriuretic peptide (NT-proBNP) and cardiac troponin I (cTnI) are now widely available in veterinary medicine at low cost. NT-proBNP serves as a marker for ventricular myocardial stretch in response to volume or pressure overload from structural heart disease, whereas cTnI becomes elevated in the bloodstream in response to myocardial damage, most commonly as a result of myocarditis, pericardial effusion, or primary myocardial disease (cardiomyopathies).

So, in today’s VETgirl podcast, we’re going to review a study by Morey et al out of University of Missouri entitled “N-terminal brain natriuretic peptide, cardiac troponin-I, and point-of-care ultrasound in dogs with cardiac and noncardiac causes of nonhemorrhagic ascites.” This study had two objectives. The primary objective was to assess the ability of biomarkers such as NT-proBNP and cTnI, along with point-of-care ultrasound (POCUS) to differentiate cardiac versus noncardiac causes of nonhemorrhagic ascites, with the suspicion that the biomarker concentrations would be higher in the cardiac group. Secondarily, they sought to compare the biomarker concentrations within the cardiac ascites group to see if the biomarker levels could differentiate between pericardial and non-pericardial disease causes, with the hypothesis that NT-proBNP would be higher in the non-pericardial disease subgroup.

This was a prospective study with recruitment over an 18-month period from 2020-2021 at a single institution. Upon confirmation of NHA via POCUS, sampling, and analysis of abdominal effusion to confirm a PCV < 10% for inclusion into this study, dogs received an additional physical examination with specific emphasis to evaluate for additional signs of cardiac disease such as jugular vein distension and/or pulsation, heart murmur, abdominal distension, and abdominal fluid wave. Each dog then underwent a complete echocardiogram, blood pressure assessment, complete blood count, serum biochemistry, plasma NT-proBNP, and serum cTnI within 12 hours of admission. If pulmonary hypertension was identified, which can be caused by either primary cardiac or a primary respiratory pathologies, echocardiogram was used to assess for the presence or absence of cardiac structural changes that would support development of pulmonary hypertension from cardiac disease.

The POCUS exam was performed according to previously described methods by a cardiologist or resident under direct supervision by a cardiologist, and evaluated offline by two blinded investigators. The POCUS evaluation documented the presence or absence of cavity effusions, subjective assessment of hepatic vein and caudal vena cava size (distended or normal), and presence or absence of gall bladder wall edema.

In this study, venous NT-proBNP and cTnI samples were drawn at time of identification of NHA, before a pericardiocentesis was performed if applicable to the case, and within 12 hours of the devised treatment plans for each patient. Samples were submitted to outside laboratories and thus were not bedside/point-of-care testing.

Based on test results, the causes of NHA were initially divided into cardiogenic nonhemorrhagic ascites (cNHA) and noncardiogenic hemorrhagic ascites (ncNHA). The cardiac NHA group was then subdivided into dogs with or without pericardial disease. All dogs with NHA were included in the analysis of physical examination and POCUS findings, but dogs were then excluded from cardiac biomarker analysis if they were deemed to be hypertensive (via Doppler-derived systolic blood pressure > 160 mm Hg), azotemic (defined as creatinine > 2.9 mg/dL), or anemia (hematocrit < 32%) due to their potential effects on NT-proBNP. Little VETgirl side note that you must be aware of? Many of our patients may present to us on ER with various diseases that produce a pre-renal azotemia; so keep in mind that azotemia is suspected to affect pro-BNP snap tests to some degree, with the 2.9mg/dL being a reported cut off in CATS which corresponds with an IRIS stage 3 in cats.

While we don’t delve into stats much during these VETgirl podcasts, these authors did perform power calculations as part of the statistical analysis, namely, to determine the minimum number of dogs needed for the biomarkers to detect a statistically significant difference. For NT-proBNP the number needed to meet was 20 dogs. However, the authors note this sample of 20 was not likely to detect differences for cTnI based on prior studies.

60 dogs met inclusion criteria for the study, including 32 with noncardiac NHA and 28 with cardiac NHA. Five dogs within the noncardiac NHA group had no cause identified whatsoever, probably due to the limited diagnostic testing performed in those patients. Within the cardiac NHA group, 17 were considered non-pericardial cNHA, thus dogs with right sided congestive heart failure due to structural right heart disease or pulmonary hypertension. Eleven were determined to have pericardial disease as the cause of their NHA. 54/60 dogs had biomarker analysis and 56/60 had POCUS. No difference was detected in mean age or body weight between the cardiac NHA and noncardiac NHA group. The cardiac NHA group did have a significantly higher proportion (33%) of males vs the noncardiac NHA group (23.3%).

With regards to physical examination, it was no surprise that dogs in the non-pericardial cardiac NHA group had significantly greater occurrence of heart murmurs than those in the pericardial group. This would be expected as the former group will include dogs with right sided valvular disease (e.g. stenosis or regurgitation), myocardial disease (with secondary tricuspid valve regurgitation), or pulmonary hypertension (with associated tricuspid regurgitation). 53.5% of dogs in the cardiac NHA group had jugular pulsation and distension compared with only 6% in the noncardiac NHA group, which was significant between groups. This would be an expected finding as dogs with cardiogenic ascites will presumably have elevated central venous pressures, for which jugular pulsation/distensions serves as a surrogate marker.

When the cardiac group was further analyzed, 65% of the dogs with non-pericardial disease had jugular pulsation and distension compared with only 36% of the pericardial group which was also a significant difference. The genesis of ascites in pericardial disease, which often is due to cardiac tamponade, still occurs as a result of elevated central venous pressures, but these patients are usually euvolemic (or hypovolemic) compared with dogs with chronic right sided congestive heart failure who instead are experiencing chronic volume overload, and so are more likely to display jugular pulsation or distension as a resultant effect of their volume overload.

The results of POCUS analysis indicated that significantly more dogs (67%) in the cardiac NHA group had hepatic vein distension compared with the noncardiac NHA group (10%). Similarly, 48% of dogs in the cardiac NHA group had caudal vena cava distension compared with only 7% in the noncardiac NHA group. Again, no surprises here, as hepatic vein and caudal vena cava distension are simply the caudal body version of jugular vein distension or pulsation – they all represent an identifiable marker for elevated central venous pressure. Even less surprising was that dogs in the cardiac NHA group were more likely to have pericardial or pleural effusion, too– funny how that works out!

Median NT-proBNP concentration was significantly higher in the cardiac NHA group (4510 pmol/L) vs the noncardiac NHA group (739.5 pmol/L, which is within the normal range). A cutoff of 4092 pmol/L carried the best combination of sensitivity and specificity, but the sensitivity was quite low (53.8%) at this cutoff, which is not surprising since a value > 900 pmol/L is considered abnormal in dogs. So at this higher cutoff value there would be a large number of false negatives occurring, or a large number of dogs with a true cardiogenic cause for their ascites would fail to be diagnosed based on pro-BNP alone. However, if the pericardial disease group is removed, the pro-BNP test sensitivity at the same cutoff improved to 81.3% for the non-pericardial cardiac NHA group, which is nothing to turn one’s nose up at. In fact, the median NT-proBNP for the pericardial group was normal in this study at 692.5 pmol/L compared with the median NT-proBNP for non-pericardial cardiac NHA at 8339 pmol/L, so the pericardial disease subgroup was dragging down the whole cardiogenic group in regards to pro-BNP. We can understand why pericardial disease would have a normal pro-BNP because many causes of pericardial disease (e.g. neoplasia, idiopathic pericarditis) typically do not have chronic structural disease with volume overload, so would not be expected to produce notable elevations in NT-proBNP. So perhaps a take home point would be that in our diagnostic workup of a non-hemorrhagic ascites, we should use POCUS to first broadly rule out pericardial disease by seeing a lack of pericardial effusion, and then we can run a pro-BNP to help differentiate whether the ascites is more likely to be from a cardiac cause or a non-cardiac cause.

Lastly, and interestingly, the cardiac troponin I levels did not differ between the cardiac and noncardiac NHA groups, or the pericardial and non-pericardial cardiac NHA groups. This can be accounted for by a variety of reasons. Even in dogs with noncardiac critical illness, cardiac troponin I is often elevated due to myocardial inflammation associated with systemic disease. In pericardial disease, myocardial ischemia, pericarditis, or inflammation associated with destruction of the myocardium by neoplasia can elevate troponin I levels. And finally, with primary structural heart disease, baseline levels of myocardial inflammation do occur which will elevate troponin I. Thus for nonhemorrhagic ascites, circulating cardiac troponin I would not necessarily serve as a very good differentiator between cardiac and non-cardiac causes when compared with NT-proBNP, as shown in this study.

What where the limitations of this study? Major limitations in this study included small sample size, study design as a two-gate study, where multiple inclusion criteria are used to create groups, and the fact that patients did receive various treatments such as fluids or medications like furosemide prior to obtaining blood samples, which could have had an effect on NT-proBNP levels.

Overall, this study was well designed and I think there are some key findings to be taken from it. First and foremost, although it is easy in this day and age to get enticed by advanced diagnostic testing, never underestimate the importance of your physical exam. This study confirmed that the physical examination alone, specifically the presence of jugular pulsation and distension, can help differentiate cardiogenic ascites from noncardiogenic ascites in a large proportion of dogs. With echocardiography and diagnostic abdominal ultrasound often immediately available in so many of our clinics, there can be a knee jerk reflex to schedule patients for a bicavitary ultrasound simply because non-hemorrhagic effusion is identified. This is premature and will spend some clients’ finances unnecessarily. If the dog with ascites has jugular vein pulsation and/or distension, the odds of cardiogenic ascites are significantly increased and a cardiac workup should be prioritized.

Secondly, if your physical examination doesn’t provide enough insight, biomarker testing – particularly NT-proBNP – and point-of-care ultrasound are your friends. They are particularly helpful as complementary tests, and in a setting where you don’t have immediate access to complete diagnostic echocardiography or abdominal ultrasound. It’s important to note the authors in this study did not use in-house canine NT-proBNP testing – the samples were sent to an outside laboratory, which will mean a few days delay for results. Thankfully, in most cases of nonhemorrhagic ascites, a few days delay isn’t usually of significant consequence to the patient, with the major exception of pericardial disease. This is where having POCUS available is so important, so that you can screen for pericardial effusion which might require emergent intervention such as pericardiocentesis. Your physical examination is also critical here, in assessing for signs of cardiac tamponade, but POCUS would be confirmatory. If practical for your clinic, you can consider purchasing an in-house NT-proBNP machine, as well.

My third take home point from this study is the affirmation that while pericardial disease and structural right heart disease are both considered “cardiac” disease capable of producing ascites, that is where the similarities end. Structural right heart disease causes ascites via volume overload due to chronic activation of the renin-angiotensin-aldosterone system (RAAS). Pericardial disease causes ascites via acute or subacute venous congestion secondary to cardiac tamponade. Dogs with pericardial causes of nonhemorrhagic ascites did not even have elevated NT-proBNP levels, on average, in this study. So, what can we take away from this VETgirl podcast? Perhaps the take home message here should be that biomarkers and point-of-care ultrasound CAN help us differentiate among the THREE general categories of nonhemorrhagic ascites: pericardial disease, right heart failure from structural heart disease, and noncardiogenic ascites.

cTnI: Cardiac troponin I
POCUS: Point-of-care ultrasound
RAAS: Renin-angiotensin-aldosterone system
NHA: Nonhemorrhagic ascites
NT-proBNP: N-terminal probrain natriuretic peptide

Morey AG, Lamb KE, Karnia JJ et al. N-terminal brain natriuretic peptide, cardiac troponin-I, and point-of-care ultrasound in dogs with cardiac and noncardiac causes of nonhemorrhagic ascites. J Vet Intern Med. 2023;37:900–909.

Today’s VETgirl podcast is sponsored by Royal Canin. Have a patient with acute or chronic gastrointestinal issues? At Royal Canin, our diets include tailored omega-3 fatty acids and a robust fiber blend to support a healthy GI microbiome. From core gastrointestinal diets to low fat, hydrolyzed protein, and fiber-enhanced options, Royal Canin’s extensive GI portfolio provides highly palatable and digestible diets to help support your patient’s GI needs! Visit my.royalcanin.com/vet/gi to learn more.

RCLogo_Tagline_2017 newest


Only VETgirl members can leave comments. Sign In or Join VETgirl now!