Assessment of serum N-terminal pro-B-type natriuretic peptide concentration for differentiation of congestive heart failure from primary respiratory tract disease as the cause of respiratory signs in dogs

Mark A. Oyama Department of Clinical Studies-Philadelphia, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104.

Search for other papers by Mark A. Oyama in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
John E. Rush Department of Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA 01536.

Search for other papers by John E. Rush in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM, DACVECC
,
Elizabeth A. Rozanski Department of Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University, North Grafton, MA 01536.

Search for other papers by Elizabeth A. Rozanski in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM, DACVECC
,
Philip R. Fox The Animal Medical Center, 510 E 62nd St, New York, NY 10021.

Search for other papers by Philip R. Fox in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Caryn A. Reynolds Department of Clinical Studies-Philadelphia, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA 19104.

Search for other papers by Caryn A. Reynolds in
Current site
Google Scholar
PubMed
Close
 DVM
,
Sonya G. Gordon Department of Small Animal Clinical Sciences, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843.

Search for other papers by Sonya G. Gordon in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Barret J. Bulmer Department of Clinical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, OR 97331.

Search for other papers by Barret J. Bulmer in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Bonnie K. Lefbom The Hope Center, 416 Maple Ave W, Vienna, VA 22151.

Search for other papers by Bonnie K. Lefbom in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Bill A. Brown Veterinary Cardiology Consultants, Animal Emergency Center, 24255 Novi Rd, Novi, MI 48375.

Search for other papers by Bill A. Brown in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Linda B. Lehmkuhl MedVet Medical & Cancer Center for Pets, 300 E Wilson Bridge Rd, Worthington, OH 43085.

Search for other papers by Linda B. Lehmkuhl in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Robert Prosek Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, FL 32608.

Search for other papers by Robert Prosek in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Michael B. Lesser Advanced Veterinary Care Center, 15926 Hawthorne Blvd, Lawndale, CA 90260.

Search for other papers by Michael B. Lesser in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Marc S. Kraus Department of Clinical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853.

Search for other papers by Marc S. Kraus in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Maribeth J. Bossbaly Heartsound Consultants, 1900 W Old Lincoln Hwy, Langhorne, PA 19047.

Search for other papers by Maribeth J. Bossbaly in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
,
Gregg S. Rapoport Angell Animal Medical Center, 350 S Huntington Ave, Boston, MA 02130.

Search for other papers by Gregg S. Rapoport in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM
, and
Jean-Sebastien Boileau Fifth Avenue Veterinary Specialists, 1 W 15th St, New York, NY 10011.

Search for other papers by Jean-Sebastien Boileau in
Current site
Google Scholar
PubMed
Close
 DVM, DACVIM

Abstract

Objective—To determine whether serum N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentration is useful in discriminating between cardiac and noncardiac (ie, primary respiratory tract disease) causes of respiratory signs (ie, coughing, stertor, stridor, excessive panting, increased respiratory effort, tachypnea, or overt respiratory distress) in dogs.

Design—Multicenter cross-sectional study.

Animals—115 dogs with respiratory signs.

Procedures—Dogs with respiratory signs were solicited for study. Physical examination, thoracic radiography, and echocardiography were used to determine whether respiratory signs were the result of cardiac (ie, congestive heart failure) or noncardiac (ie, primary respiratory tract disease) causes. Serum samples for NT-proBNP assay were obtained at time of admission for each dog. Receiver-operating characteristic curves were constructed to determine the ability of serum NT-proBNP concentration to discriminate between cardiac and noncardiac causes of respiratory signs.

Results—Serum NT-proBNP concentration was significantly higher in dogs with cardiac versus noncardiac causes of respiratory signs. In dogs with primary respiratory tract disease, serum NT-proBNP concentration was significantly higher in those with concurrent pulmonary hypertension than in those without. A serum NT-proBNP cutoff concentration > 1,158 pmol/L discriminated between dogs with congestive heart failure and dogs with primary respiratory tract disease with a sensitivity of 85.5% and a specificity of 81.3%.

Conclusions and Clinical Relevance—Measuring serum NT-proBNP concentration in dogs with respiratory signs helps to differentiate between congestive heart failure and primary respiratory tract disease as an underlying cause.

Abstract

Objective—To determine whether serum N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentration is useful in discriminating between cardiac and noncardiac (ie, primary respiratory tract disease) causes of respiratory signs (ie, coughing, stertor, stridor, excessive panting, increased respiratory effort, tachypnea, or overt respiratory distress) in dogs.

Design—Multicenter cross-sectional study.

Animals—115 dogs with respiratory signs.

Procedures—Dogs with respiratory signs were solicited for study. Physical examination, thoracic radiography, and echocardiography were used to determine whether respiratory signs were the result of cardiac (ie, congestive heart failure) or noncardiac (ie, primary respiratory tract disease) causes. Serum samples for NT-proBNP assay were obtained at time of admission for each dog. Receiver-operating characteristic curves were constructed to determine the ability of serum NT-proBNP concentration to discriminate between cardiac and noncardiac causes of respiratory signs.

Results—Serum NT-proBNP concentration was significantly higher in dogs with cardiac versus noncardiac causes of respiratory signs. In dogs with primary respiratory tract disease, serum NT-proBNP concentration was significantly higher in those with concurrent pulmonary hypertension than in those without. A serum NT-proBNP cutoff concentration > 1,158 pmol/L discriminated between dogs with congestive heart failure and dogs with primary respiratory tract disease with a sensitivity of 85.5% and a specificity of 81.3%.

Conclusions and Clinical Relevance—Measuring serum NT-proBNP concentration in dogs with respiratory signs helps to differentiate between congestive heart failure and primary respiratory tract disease as an underlying cause.

Differentiating between causes of respiratory distress in animals that have acute signs is challenging. Clinical signs are generally nonspecific and can result from congestive heart failure as well as primary respiratory tract disease. Data gathered through medical history, physical examination, and thoracic radiography can be ambiguous or nonconfirmatory. Differentiation of the common causes of respiratory signs (ie, coughing, stertor, stridor, excessive panting, increased respiratory effort, tachypnea, or overt respiratory distress) is essential to optimize management and outcome. Thus, a blood-based assay to differentiate between cardiac versus noncardiac (ie, primary respiratory tract disease) causes of respiratory signs would be useful.

N-terminal pro-B-type natriuretic peptide is a cardiac biomarker that is increased in the blood of dogs with heart disease.1,2 N-terminal pro-B-type natriuretic peptide is formed when its parent pro-hormone, pro-B-type natriuretic peptide, is cleaved into 2 molecules, NT-proBNP and C-BNP. The pro-B-type natriuretic peptide is produced in response to an increase in intracardiac hydrostatic pressure, increased cardiac wall stress, angiotensin II, myocardial hypoxia, and heightened sympathetic tone.3 As such, both serum NT-proBNP and C-BNP concentrations represent potential diagnostic tools to help diagnose congestive heart failure.4,5 In humans, C-BNP6–8 and NT-proBNP9–11 assays possess high sensitivity and specificity in differentiating between cardiac and pulmonary causes of respiratory distress. In circulation, C-BNP is rapidly degraded,12 making laboratory testing difficult. Commercially available tests in dogs and cats exclusively detect NT-proBNP, which is thought to possess a longer half-life.

Results of a previous study1 indicate that dogs with respiratory signs resulting from congestive heart failure have significantly higher serum NT-proBNP concentrations than dogs with primary respiratory tract disease. The magnitude of difference between the 2 populations was such that the serum NT-proBNP concentration provided a clinically useful test with an adequate sensitivity and specificity for differentiating between the 2 populations. However, all dogs did not undergo the same set of diagnostic tests, and a substantial proportion of the study population lacked both evaluation of thoracic radiographs and an echocardiographic examination. Thus, the usefulness of this peptide as an adjunct diagnostic test merits closer evaluation. We hypothesized that serum NT-proBNP concentration would be higher in dogs with congestive heart failure than in dogs with primary respiratory tract disease and that serum NT-proBNP concentration would help to differentiate between cardiac and noncardiac causes of respiratory signs in dogs. The purpose of the multicenter study reported here was to assess the clinical sensitivity and specificity of a serum NT-proBNP assay to differentiate between cardiac versus noncardiac (ie, primary respiratory tract disease) causes of respiratory signs, and to determine an optimal cutoff value that would facilitate this assessment.

Materials and Methods

Animals—Study procedures were approved by the institutional animal use and care committees of each qualifying participating institution. Owner consent was obtained for all dogs included in the study. Fourteen veterinary cardiology practices prospectively recruited dogs between September 2006 and November 2007. Dogs were eligible for inclusion if the owner reported a complaint of respiratory signs that were severe enough to affect the dog's quality of life. Qualifying signs included coughing, stertor, stridor, excessive panting, increased respiratory effort, tachypnea, or overt respiratory distress. Dogs were excluded if respiratory signs were caused by obvious trauma (eg, vehicular trauma).

Assessment of disease—All dogs underwent thoracic radiography and M-mode, 2-D, and Doppler echocardiography. Vertebral heart size was measured on the left or right lateral radiographic projection of the thorax by the attending cardiologist as previously described.13,14 Additional diagnostic testing to help determine the underlying cause of the respiratory signs (eg, fluoroscopy, tracheal wash, and bronchoscopy) was performed at the discretion of the attending clinician. The LVIDd, LVIDs, and LA:Ao were measured by means of standard echocardiographic techniques.15,16 Detection of tricuspid regurgitation by use of echocardiography was recorded. Pulmonary hypertension was arbitrarily defined as a velocity of tricuspid regurgitation > 3.25 m/s (pressure gradient > 42.3 mm Hg). In the absence of tricuspid regurgitation, a presumptive echocardiographic diagnosis of pulmonary hypertension was made in dogs with a combination of severe right ventricular hypertrophy or dilation, flattened interventricular septum, and an enlarged main pulmonary artery, in the absence of increased right ventricular to pulmonary artery outflow velocity.

A standardized patient datasheet was used to record patient signalment, clinical signs, and diagnostic findings. On the basis of this workup, and without knowledge of the NT-proBNP assay results, a board-certified veterinary cardiologist assigned the dog to 1 of 4 groups as follows: group 1 included dogs with congestive heart failure (findings of severe cardiac disease, perihilar or caudodorsal pulmonary interstitial pattern, cardiomegaly, enlarged pulmonary veins, pleural effusion, and ascites) and without primary respiratory tract disease; group 2 included dogs with primary respiratory tract disease (parenchymal lung disease [findings of cranioventral or lobar alveolar pattern, lung lobe torsion, and bullae] or upper airway disease [findings of laryngeal paralysis, brachycephalic syndrome, collapsing trachea, and moderate or severe bronchiolar pulmonary pattern]) and without underlying cardiac disease (no heart murmur and no evidence of clinically relevant left-sided cardiac disease on echocardiography); group 3 included dogs with primary respiratory tract disease (as described for group 2 dogs) that also had heart disease that was not congestive heart failure (detection of heart murmur or cardiac disease on echocardiography without evidence of congestive heart failure on thoracic radiography, physical examination, or other diagnostic tests); and group 4 included dogs for which the etiology of respiratory signs could not be reliably ascribed to either respiratory tract disease or congestive heart failure.

Measurement of serum NT-proBNP concentration—Venous blood samples were collected at the time of admission. Blood was drawn into plain evacuated glass tubes that did not contain any additives. Samples were centrifuged within 60 minutes after collection, and serum was stored at −20°C prior to batched overnight shipment for analysis. Samples were shipped with cold packs and packing materials provided by the assay manufacturer. Serum NT-proBNP concentration was determined with a commercially available canine-specific NT-proBNP assaya as previously described.2

Statistical analysis—Data were summarized as median (IQR [ie, 25th to 75th percentile]) or mean ± SD. Multiple group comparisons of radiographic, echocardiographic, and NT-proBNP data were performed via Kruskal-Wallis analysis and the Dunn multiple comparison tests. The Spearman method was used to test for correlations between serum NT-proBNP concentration and body weight, age, and radiographic and echocardiographic heart size. Multivariate linear regression was performed to identify whether these variables were associated with serum NT-proBNP concentration; cutoff P values for entry into and removal from the multivariate model were < 0.05 and > 0.10, respectively. Receiver-operating characteristic curves were constructed to determine the sensitivity, specificity, positive and negative predictive values, accuracy (defined as the sum of the concordant results divided by the sum of all results in a 2 × 2 table), and AUCROC of NT-proBNP results, compared with the cardiologist's diagnosis. All analyses were performed with standard software.b,c Values of P < 0.05 were considered significant.

Results

Animals—One hundred ninety-six dogs from 14 sites were entered into the study (University of Pennsylvania, n = 59 dogs; Tufts University, 59; The Animal Medical Center, 11; Oregon State University, 10; Texas A&M University, 9; Chesapeake Veterinary Cardiology Associates, 8; Veterinary Cardiology Consultants, 8; Advanced Veterinary Care Center, 6; Angell Animal Medical Center, 6; MedVet Medical & Cancer Center for Pets, 6; University of Florida, 5; Cornell University, 4; Fifth Avenue Veterinary Specialists, 3; Heartsound Consultants, 2). Complications involving shipping of serum samples resulted in the loss of the first 81 samples obtained. These samples were shipped according to original manufacturer recommendations but arrived at the laboratory at room temperature (approx 23°C). Following this incident, the manufacturer changed its shipping recommendations to ensure transport and overnight arrival at 4°C. This involved use of extra cold packs and heavily insulated shipping containers, and the following 115 samples shipped in this fashion were used in the study analysis. Fifteen of the 115 (13%) dogs were of mixed breeding, 9 (8%) were Cavalier King Charles Spaniels, 6 (5%) were Shih Tzus, 5 (4%) were Cocker Spaniels, and 5 (4%) were Chihuahuas; the remaining dogs represented 41 other breeds. Forty (35%) dogs were spayed females, 4 (3%) were sexually intact females, 61 (53%) were neutered males, and 10 (9%) were sexually intact males. Clinical signs of respiratory problems were recorded for the 196 originally enrolled dogs as follows: 76 instances of coughing, 46 instances of moderate tachypnea or dyspnea, 27 instances of mild tachypnea or dyspnea, 24 instances of excessive panting, 19 instances of severe tachypnea or dyspnea, and 4 instances of stertor or stridor. Thirty-seven of the 115 (32%) dogs had only 1 clinical sign, 69 (60%) had 2 clinical signs, and 9 (8%) had ≥ 3 clinical signs.

Etiology of respiratory signs—On the basis of findings on physical examination, thoracic radiography, and echocardiography, 62 of the 115 (54%) dogs were placed into group 1, 21 (18%) into group 2, 27 (23%) into group 3, and 5 (4%) into group 4. For the 62 group 1 dogs, underlying causes of congestive heart failure included chronic degenerative mitral valve disease (44/62 [71%]), dilated cardiomyopathy (13/62 [21%]), congenital heart disease (2/62 [3%]), endocarditis (1/62 [2%]), heart block (1/62 [2%]), and arrhythmogenic right ventricular cardiomyopathy (1/62 [2%]). For the 21 group 2 dogs, primary respiratory tract diseases included small airway disease (8/21 [38%]), upper airway disease including collapsing trachea (5/21 [24%]), thoracic or pulmonary neoplasia (3/21 [14%]), pneumonia (3/21 [14%]), lung lobe torsion (1/21 [5%]), and noncardiogenic pulmonary edema (1/21 [5%]). For the 27 group 3 dogs, primary respiratory tract diseases included small airway disease (12/27 [44%]), upper airway disease including collapsing trachea (8/27 [30%]), pneumonia (4/27 [15%]), pulmonary or thoracic neoplasia (2/27 [7%]), and suspected pulmonary thromboembolism (1/27 [4%]). For the 27 group 3 dogs, concurrent underlying heart diseases included degenerative mitral valve disease (25/27 [93%]), severe tricuspid valve disease (1/27 [4%]), and left ventricular hypertrophy resulting from systemic hypertension (1/27 [4%]). Group 4 included 5 dogs with various combinations of severe pulmonary hypertension, parenchymal pulmonary disease, severe degenerative valvular disease, pericardial effusion, myocardial failure, and ventricular arrhythmias. Because of the inability to definitively diagnose the etiology of respiratory signs in group 4 dogs, they were excluded from further analysis.

Vertebral heart size, LVIDd, and LVIDs indexed to body weight were significantly different among groups. Median (IQR) VHS was 12.3 (11.5 to 13.0), 10.5 (10.0 to 11.0), and 11.0 (10.0 to 11.5) for group 1, 2, and 3 dogs (P < 0.001), respectively. Median VHS was significantly different between group 1 and 2 dogs (P < 0.001) as well as between group 1 and 3 dogs (P < 0.001). Median (IQR) LVIDd indexed to body weight was 0.355 (0.216 to 0.534), 0.214 (0.136 to 0.317), and 0.303 (0.231 to 0.451) for group 1, 2, and 3 dogs (P = 0.005), respectively. Median LVIDd was significantly different between group 1 and 2 dogs (P = 0.001) as well as between group 2 and 3 dogs (P = 0.015). Median LVIDs indexed to body weight was 0.196 (0.151 to 0.283), 0.095 (0.078 to 0.168), and 0.183 (0.135 to 0.235) for group 1, 2, and 3 dogs (P < 0.001), respectively. Median LVIDs indexed to body weight was significantly different between group 1 and 2 dogs (P < 0.001) as well as between group 2 and 3 dogs (P = 0.003). Median (IQR) LA:Ao was 2.24 (1.90 to 2.79), 1.29 (1.15 to 1.49), and 1.40 (1.19 to 1.70) for group 1, 2, and 3 dogs (P < 0.001), respectively. Median LA:Ao was significantly different between group 1 and 2 dogs (P < 0.001) as well as between group 1 and 3 dogs (P < 0.001).

Serum NT-proBNP concentrations—Median serum NT-proBNP concentration was significantly (P < 0.001) different among groups. Median (IQR) serum NT-proBNP concentration of group 1 dogs (2,445 pmol/L [1,499 to 3,134 pmol/L]) was significantly higher than in group 2 (413 pmol/L [245 to 857 pmol/L]; P < 0.001) and group 3 (478 pmol/L [323 to 1,158 pmol/L]; P < 0.001) dogs. Median serum NT-proBNP concentration between group 2 and 3 dogs was not significantly different (Figure 1).

Figure 1—
Figure 1—

Box and whisker plot of serum NT-proBNP concentration in dogs in which the etiology of respiratory signs is congestive heart failure (group 1; n = 62), primary respiratory tract disease (group 2; 21), and respiratory tract disease with concurrent heart disease (group 3; 27). For each plot, the box represents the IQR, the horizontal line in the middle of the box represents the median, and the whiskers denote the range extending to 1.5 times the IQR from the upper and lower quartiles. Outlier values between 1.5 to 3.0 times the IQR are denoted as squares. *Significantly (P < 0.005) different from value for group 1 dogs.

Citation: Journal of the American Veterinary Medical Association 235, 11; 10.2460/javma.235.11.1319

Tricuspid regurgitation was detected in 34 (34/62 [55%]), 5 (5/21 [24%]), and 18 (18/27 [67%]) group 1, 2, and 3 dogs, respectively. A diagnosis of pulmonary hypertension was made for 16 (16/62 [26%]) group 1 dogs (mean ± SD tricuspid regurgitation velocity, 3.78 ± 0.49 m/s; n = 16), 7 (7/21 [33%]) group 2 dogs (4.69 ± 0.89 m/s; 5), and 5 (5/27 [19%]) group 3 dogs (4.27 ± 0.60 m/s; 5). Median (IQR) serum NT-proBNP concentration in group 2 dogs with pulmonary hypertension was significantly (P = 0.002) higher than in group 2 dogs without pulmonary hypertension (1,028 pmol/L [559 to 1,829 pmol/L] vs 309 pmol/L [196 to 461 pmol/L], respectively). In contrast, median (IQR) serum NT-proBNP concentration in group 1 dogs with pulmonary hypertension was not significantly (P = 0.41) higher than in group 1 dogs without pulmonary hypertension (2,078 pmol/L [1,484 to 3,071 pmol/L] vs 2,541 pmol/L [1,539 to 3,141 pmol/L], respectively). Also, median (IQR) serum NT-proBNP concentration in group 3 dogs with pulmonary hypertension was not significantly (P = 0.81) higher than in group 3 dogs without pulmonary hypertension (478 pmol/L [348 to 1,616 pmol/L] vs 504 pmol/L [301 to 1,107 pmol/L], respectively).

Univariate regression analysis revealed that serum NT-proBNP concentration was significantly but poorly correlated to body weight (r = 0.21; P = 0.029), VHS (r = 0.55; P < 0.001), LVIDs indexed to body weight (r = 0.27; P = 0.021), and LA:Ao (r = 0.61; P < 0.001). Multiple regression analysis revealed that only LA:Ao (β coefficient = 757.9; P < 0.001) and VHS (β coefficient = 218.3; P = 0.043) were correlated with serum NT-proBNP concentration.

Receiver-operating characteristic curve analysis—Serum NT-proBNP concentrations > 1,158 pmol/L differentiated group 1 dogs from group 2 and 3 dogs with a sensitivity of 85.5%, specificity of 81.3%, positive predictive value of 85.5%, negative predictive value of 81.3%, and accuracy of 83.6%. The AUCROC was 90.5% (95% CI, 83.4% to 95.2%; Figure 2). Additional cutoff values for serum NT-proBNP that yielded either a sensitivity > 90% or a specificity > 90% were identified (NT-proBNP > 963 pmol/L [sensitivity, 90.3%; specificity, 73.5%]; NT-proBNP > 1,829 pmol/L [sensitivity, 64.5%; specificity, 91.7%]). The AUCROC was 93.0% (95% CI, 85.2% to 97.4%) when differentiating between group 1 dogs and group 2 dogs and was 88.5% (95% CI, 79.9% to 94.3%) when differentiating between group 1 dogs and group 3 dogs.

Figure 2—
Figure 2—

The ROC curve displaying the sensitivity and specificity of serum NT-proBNP concentration to distinguish between cardiac and noncardiac (ie, primary respiratory tract disease) as the cause of respiratory signs in 110 dogs (solid line). The 95% CIs are displayed as the dashed lines. The diagonal dotted line represents the line of no discrimination. Various potential diagnostic cutoff values (pmol/L) are indicated along the curve. AUCROC = 0.905.

Citation: Journal of the American Veterinary Medical Association 235, 11; 10.2460/javma.235.11.1319

Discussion

Our results indicate that serum NT-proBNP concentration in dogs with respiratory signs helps to differentiate between congestive heart failure and primary respiratory tract disease as an underlying cause. Our results are consistent with those of other studies1,17–19,d that have evaluated circulating C-BNP and NT-proBNP concentrations in dogs with respiratory tract disease. DeFrancesco et al17 reported that plasma C-BNP concentration had a sensitivity of 90% and a specificity of 78% in differentiating between 101 dogs with congestive heart failure and 78 dogs with respiratory tract disease. Prosek et al18 reported that plasma C-BNP concentration had a sensitivity of 86.4% and a specificity of 80.8% in differentiating between 22 dogs with congestive heart failure and 26 dogs with respiratory tract disease. Wess et ald reported that a plasma NT-proBNP concentration > 520 pmol/L had a sensitivity of 94.7% and a specificity of 96.2% in differentiating between 19 dogs with cardiac disease and 57 dogs with respiratory tract disease. In a study of 46 dogs, Fine et al19 reported that a serum or plasma NT-proBNP concentration > 1,400 pmol/L detected 92% of dogs with heart failure. Boswood et al1 reported that a plasma NT-proBNP concentration > 210 pmol/L had a sensitivity of 85.5% and a specificity of 82.4% in differentiating between dogs with cardiac disease and those with respiratory tract disease. Thus, our results are in general agreement with those of other studies. Our study was unique in its large multicentered design and the requirement that all dogs undergo both thoracic radiography and echocardiography as part of their diagnostic workup. In our study as well as in others in humans20–22 and dogs,2,23 serum NT-proBNP concentration was correlated with echocardiographic and radiographic measures of cardiac enlargement.

Our results as well as those of Fine et al19 indicate a higher diagnostic cutoff value for serum NT-proBNP concentration than those of studies by Boswood et al1 and Wess et ald; the reason for the discrepancy is unclear. Results of previous studies have revealed ranges of circulating NT-proBNP concentrations in both healthy dogs2,24 and dogs with heart disease or respiratory tract disease2,24,25 that are higher than those initially reported by Boswood et al.1 It is interesting to note that the lower circulating NT-proBNP concentrations were reported from studies1,d in Europe, and differences in the study population or sample handling may be responsible for the variation in values. In humans, NT-proBNP, while relatively stable, compared with C-BNP, degrades if samples are left at room temperature,26 and careful attention to sample collection, handling, and shipping is required. To the authors' knowledge, the in vitro stability of canine NT-proBNP has not been reported; however, our experience with shipping protocols indicates that NT-proBNP degrades during overnight shipping if not maintained at 4°C. Future studies are needed to further evaluate different handling and shipping protocols before wide-ranging clinical recommendations regarding NT-proBNP assay results can be made.

The clinical usefulness of diagnostic assays can be compared among various studies27 by examining results of ROC curve analysis. Inspection of the ROC curve indicates assay sensitivity and specificity at any given cutoff value. The AUCROC reflects the discriminatory ability of the test with 50% indicating no ability and 100% indicating perfect ability to discriminate between conditions. To be clinically useful, diagnostic tests should possess an AUCROC of at least 75% to 85%.27 Our study yielded an AUCROC of 90.5%, which is comparable to results from 3 of the largest studies9,10,28 in humans. Maisel et al28 reported that plasma or whole blood C-BNP concentration had a 90% sensitivity, a 76% specificity, and an AUCROC of 91% for predicting the presence or absence of congestive heart failure in a cohort of 1,586 patients with shortness of breath. Januzzi et al9 reported that a circulating NT-proBNP concentration > 900 pg/mL had an 87% sensitivity, an 86% specificity, and an AUCROC of 94% for predicting acute congestive heart failure in 600 patients with dyspnea, while Moe et al10 reported that, compared with traditional evaluation without the use of biomarker assays, use of a NT-proBNP assay increased the AUCROC from 83% to 90% for predicting acute heart failure in 500 patients presenting to the emergency room with dyspnea. In humans, natriuretic peptide testing is currently recommended as one of the first clinical steps in evaluating patients suspected of having heart failure.29 Use of NT-proBNP assay in these patients is strongly associated with shortened hospital stay, lower rate of rehospitalization, and reduced financial cost.8,10

In our study as well as in others,9,11,28,30 the diagnostic gold standard was the clinical evaluation by a board-certified cardiologist of a uniform set of diagnostic tests, including physical examination, thoracic radiography, and echocardiography. Our study included a diagnostic category for instances where a definitive clinical diagnosis could not be confidently made (group 4), and it is unlikely that large numbers of misdiagnoses were made in groups 1 through 3. Despite this protocol, differentiation of cardiac versus respiratory tract disease in dogs can be challenging, and it is possible that misdiagnoses were made in our study, especially in dogs with concurrent heart and respiratory tract disease. If so, this may have affected the sensitivity and specificity that were determined for the NT-proBNP assay. One potential limitation of our study involves the use of secondary and tertiary referral centers from which patients were recruited. This may introduce bias into the study's patient population; however, workup by cardiology specialists performing echocardiography was an integral part of the study and necessitated use of this particular patient population.

A diagnostic assay has potential clinical value if the current means of diagnosis possesses accuracy lower than that afforded by the assay. In our study, the accuracy of NT-proBNP assay was 83.6%. In humans, the accuracy of correctly diagnosing congestive heart failure by emergency room physicians is surprisingly low, ranging from 60% to 75%, and missed diagnoses contribute to the in-hospital mortality rate.27,31 In one of the largest studies,7 involving 1,538 patients admitted to the emergency room with congestive heart failure, accuracy of diagnosis by the primary physician was only 74%. This accuracy increased to 81.5% if circulating NT-proBNP concentrations were used in conjunction with the primary physician's assessment. To our knowledge, the accuracy of diagnosing congestive heart failure in dogs by veterinarians in private practice has not been reported. This variable should be explored against NT-proBNP–assisted workup in future studies.

The clinical performance of a diagnostic test is a competing balance between sensitivity and specificity. Gains in sensitivity are made at the expense of specificity and vice versa. The ROC curve analysis allows identification of various cutoff values, which maximize either parameter. In our study, lowering the cutoff value to 963 pmol/L increased sensitivity to > 90%, while increasing the cutoff value to 1,829 pmol/L increased specificity to > 90%. In previous studies in dogs2 and humans,27,27,32 an intermediate gray zone, in which assay results are less reliable, has been proposed. In a population of dogs with mitral valve disease and dilated cardiomyopathy,2 a serum NT-proBNP concentration > 1,725 pmol/L or < 820 pmol/L yielded good sensitivity and specificity for discriminating between dogs with and without congestive heart failure; however, the clinical usefulness of serum NT-proBNP concentrations between those 2 values was limited. In humans, NT-proBNP clinical algorithms use subjective diagnostic terminology, such as unlikely, possible, and very likely, to convey assay results.28,29,33 Results that fall into the possible category automatically trigger additional diagnostic investigation to increase (or decrease) the confidence of diagnosis. These algorithms underscore the diagnostic limitations of natriuretic peptide assay in both humans and dogs, and NT-proBNP assay results should complement and not replace the medical history, physical examination, and conventional diagnostic testing.

In our study, most of the patient population comprised geriatric small-breed dogs, and as such, many dogs had concurrent cardiac and respiratory tract disease or pulmonary hypertension. Results of previous studies2,24 indicate that, although circulating NT-proBNP concentration is generally low in dogs with asymptomatic heart disease, it can be markedly high in individual dogs. These high concentrations would decrease the discriminatory ability of NT-proBNP in dogs with concurrent cardiac and respiratory tract disease. Indeed, in our study, the discriminatory power of NT-proBNP was lower in distinguishing between group 1 and 3 dogs versus between group 1 and 2 dogs. Our results indicate that serum NT-proBNP concentration is high in dogs with respiratory tract disease and pulmonary hypertension. In humans, NT-proBNP is produced by the right ventricle in response to myocyte stress,34,35 and although the high plasma NT-proBNP concentrations are generally lower than for diseases that primarily affect the left ventricle,36 serum NT-proBNP concentration is correlated with severity of right ventricular dysfunction and pressure overload.33–36 In our study, dogs with primary respiratory tract disease and pulmonary hypertension had a 3-fold increase in the median serum NT-proBNP concentration, compared with dogs with respiratory tract disease without pulmonary hypertension. This phenomenon may contribute to a likelihood of false-positive result for congestive heart failure (ie, the diagnosis of congestive heart failure in dogs with primary respiratory tract disease as the cause of respiratory signs). This concept of false-positive result for congestive heart failure on the basis of NT-proBNP assay results in patients with respiratory and cardiac disease is intriguing. On one hand, the production of NT-proBNP in response to pulmonary hypertension confounds the determination of the cause of respiratory signs. On the other hand, assay results in this population should not be summarily discounted, as NT-proBNP is strongly correlated to mortality and morbidity rates. In human patients with pulmonary hypertension, a serum NT-proBNP concentration > 1,400 pmol/L had a sensitivity of 100% for predicting the 3-year mortality rate and, in this respect, was superior to other parameters, such as exercise test results, pulmonary vascular resistance, functional clinical class, and cardiac index.37 Thus, NT-proBNP assay results may have important prognostic value in dyspneic patients despite their potential to confound diagnosis of etiology.

In conclusion, serum NT-proBNP concentration in dogs with respiratory signs in the present study helped to differentiate between congestive heart failure and primary respiratory tract disease as the underlying cause. The presence of concurrent cardiac and respiratory tract disease or pulmonary hypertension could confound diagnosis. The NT-proBNP assay should be used in conjunction with medical history, physical examination findings, and other diagnostic test results to help achieve a diagnosis for the cause of respiratory signs in dogs.

ABBREVIATIONS

AUCROC

Area under the receiver-operating characteristic curve

C-BNP

C-terminal B-type natriuretic peptide

CI

Confidence interval

IQR

Interquartile range

LA:Ao

Left atrial-to-aortic root dimension ratio

LVIDd

Left ventricular internal dimension at end-diastole

LVIDs

Left ventricular internal dimension at end-systole

NT-proBNP

N-terminal pro-B-type natriuretic peptide

ROC

Receiver-operating characteristic

VHS

Vertebral heart size

a.

Canine CardioCare NT-proBNP, Veterinary Diagnostics Institute, Irvine, Calif.

b.

GraphPad Prism, version 4.0, Graph Pad Software Inc, San Diego, Calif.

c.

MedCalc, version 9.5.1.0, MedCalc Software, Mariakerke, Belgium.

d.

Wess G, Timper N, Hirschberger J. The utility of NT-proBNP to differentiate cardiac and non-cardiac causes of coughing or dyspnea in dogs (abstr). J Vet Intern Med 2007;21:608.

References

  • 1.

    Boswood A, Dukes-McEwan J, Loureiro J, et al. The diagnostic accuracy of different natriuretic peptides in the investigation of canine cardiac disease. J Small Anim Pract 2008;49:2632.

    • Search Google Scholar
    • Export Citation
  • 2.

    Oyama MA, Fox PR, Rush JE, et al. Clinical utility of serum N-terminal pro-B-type natriuretic peptide concentration for identifying cardiac disease in dogs and assessing disease severity. J Am Vet Med Assoc 2008;232:14961503.

    • Search Google Scholar
    • Export Citation
  • 3.

    de Bold AJ, Ma KK, Zhang Y, et al. The physiological and pathophysiological modulation of the endocrine function of the heart. Can J Physiol Pharmacol 2001;79:705714.

    • Search Google Scholar
    • Export Citation
  • 4.

    Braunwald E. Biomarkers in heart failure. N Engl J Med 2008;358:21482159.

  • 5.

    Maisel A, Mueller C, Adams K Jr, et al. State of the art: using natriuretic peptide levels in clinical practice. Eur J Heart Fail 2008;10:824839.

    • Search Google Scholar
    • Export Citation
  • 6.

    Maisel AS. The diagnosis of acute congestive heart failure: role of BNP measurements. Heart Fail Rev 2003;8:327334.

  • 7.

    McCullough PA, Nowak RM, McCord J, et al. B-type natriuretic peptide and clinical judgment in emergency diagnosis of heart failure: analysis from Breathing Not Properly (BNP) Multinational Study. Circulation 2002;106:416422.

    • Search Google Scholar
    • Export Citation
  • 8.

    Mueller C, Scholer A, Laule-Kilian K, et al.Use of B-type natriuretic peptide in the evaluation and management of acute dyspnea. N Engl J Med 2004;350:647654.

    • Search Google Scholar
    • Export Citation
  • 9.

    Januzzi JL Jr, Camargo CA, Anwaruddin S, et al. The N-terminal Pro-BNP investigation of dyspnea in the emergency department (PRIDE) study. Am J Cardiol 2005;95:948954.

    • Search Google Scholar
    • Export Citation
  • 10.

    Moe GW, Howlett J, Januzzi JL, et al. N-terminal pro-B-type natriuretic peptide testing improves the management of patients with suspected acute heart failure: primary results of the Canadian prospective randomized multicenter IMPROVE-CHF study. Circulation 2007;115:31033110.

    • Search Google Scholar
    • Export Citation
  • 11.

    Alibay Y, Beauchet A, El Mahmoud R, et al.Plasma N-terminal probrain natriuretic peptide and brain natriuretic peptide in assessment of acute dyspnea. Biomed Pharmacother 2005;59:2024.

    • Search Google Scholar
    • Export Citation
  • 12.

    Pemberton CJ, Johnson ML, Yandle TG, et al. Deconvolution analysis of cardiac natriuretic peptides during acute volume overload. Hypertension 2000;36:355359.

    • Search Google Scholar
    • Export Citation
  • 13.

    Buchanan JW, Bücheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc 1995;206:194199.

  • 14.

    Hansson K, Haggstrom J, Kvart C, et al. Interobserver variability of vertebral heart size measurements in dogs with normal and enlarged hearts. Vet Radiol Ultrasound 2005;46:122130.

    • Search Google Scholar
    • Export Citation
  • 15.

    Hansson K, Haggstrom J, Kvart C, et al. Left atrial to aortic root indices using two-dimensional and M-mode echocardiography in cavalier King Charles spaniels with and without left atrial enlargement. Vet Radiol Ultrasound 2002;43:568575.

    • Search Google Scholar
    • Export Citation
  • 16.

    Sahn DJ, DeMaria A, Kisslo J, et al. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:10721083.

    • Search Google Scholar
    • Export Citation
  • 17.

    DeFrancesco TC, Rush JE, Rozanski EA, et al. Prospective clinical evaluation of an ELISA B-type natriuretic peptide assay in the diagnosis of congestive heart failure in dogs presenting with cough or dyspnea. J Vet Intern Med 2007;21:243250.

    • Search Google Scholar
    • Export Citation
  • 18.

    Prosek R, Sisson DD, Oyama MA, et al. Distinguishing cardiac and noncardiac dyspnea in 48 dogs using plasma atrial natriuretic factor, B-type natriuretic factor, endothelin, and cardiac troponin-I. J Vet Intern Med 2007;21:238242.

    • Search Google Scholar
    • Export Citation
  • 19.

    Fine DM, Declue AE, Reinero CR. Evaluation of circulating amino terminal-pro-B-type natriuretic peptide concentration in dogs with respiratory distress attributable to congestive heart failure or primary pulmonary disease. J Am Vet Med Assoc 2008;232:16741679.

    • Search Google Scholar
    • Export Citation
  • 20.

    Karabulut A, Kaplan A, Aslan C, et al. The association between NT-proBNP levels, functional capacity and stage in patients with heart failure. Acta Cardiol 2005;60:631638.

    • Search Google Scholar
    • Export Citation
  • 21.

    Mueller T, Gegenhuber A, Poelz W, et al. Biochemical diagnosis of impaired left ventricular ejection fraction–comparison of the diagnostic accuracy of brain natriuretic peptide (BNP) and amino terminal proBNP (NT-proBNP). Clin Chem Lab Med 2004;42:159163.

    • Search Google Scholar
    • Export Citation
  • 22.

    Silver MA, Maisel A, Yancy CW, et al. BNP Consensus Panel 2004: a clinical approach for the diagnostic, prognostic, screening, treatment monitoring, and therapeutic roles of natriuretic peptides in cardiovascular diseases. Congest Heart Fail 2004;10:130.

    • Search Google Scholar
    • Export Citation
  • 23.

    MacDonald KA, Kittleson MD, Munro C, et al. Brain natriuretic peptide concentration in dogs with heart disease and congestive heart failure. J Vet Intern Med 2003;17:172177.

    • Search Google Scholar
    • Export Citation
  • 24.

    Tarnow I, Olsen LH, Kvart C, et al. Predictive value of natriuretic peptides in dogs with mitral valve disease. Vet J 2008;180:195201.

  • 25.

    Hori Y, Tsubaki M, Katou A, et al. Evaluation of NT-pro BNP and CT-ANP as markers of concentric hypertrophy in dogs with a model of compensated aortic stenosis. J Vet Intern Med 2008;22:11181123.

    • Search Google Scholar
    • Export Citation
  • 26.

    Michaud K, Augsburger M, Donze N, et al. Evaluation of postmortem measurement of NT-proBNP as a marker for cardiac function. Int J Legal Med 2008;122:415420.

    • Search Google Scholar
    • Export Citation
  • 27.

    Ray P, Delerme S, Jourdain P, et al. Differential diagnosis of acute dyspnea: the value of ≤ natriuretic peptides in the emergency department. QJM 2008;101:831843.

    • Search Google Scholar
    • Export Citation
  • 28.

    Maisel AS, Krishnaswamy P, Nowak RM, et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med 2002;347:161167.

    • Search Google Scholar
    • Export Citation
  • 29.

    Swedberg K, Cleland J, Dargie H, et al. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26:11151140.

    • Search Google Scholar
    • Export Citation
  • 30.

    Berdague P, Caffin PY, Barazer I, et al. Use of N-terminal prohormone brain natriuretic peptide assay for etiologic diagnosis of acute dyspnea in elderly patients. Am Heart J 2006;151:690698.

    • Search Google Scholar
    • Export Citation
  • 31.

    Ray P, Birolleau S, Lefort Y, et al. Acute respiratory failure in the elderly: etiology, emergency diagnosis and prognosis. Crit Care 2006;10:R82.

    • Search Google Scholar
    • Export Citation
  • 32.

    van Kimmenade RR, Pinto YM, Bayes-Genis A, et al. Usefulness of intermediate amino-terminal pro-brain natriuretic peptide concentrations for diagnosis and prognosis of acute heart failure. Am J Cardiol 2006;98:386390.

    • Search Google Scholar
    • Export Citation
  • 33.

    Mueller C, Breidthardt T, Laule-Kilian K, et al.The integration of BNP and NT-proBNP into clinical medicine. Swiss Med Wkly 2007;137:412.

    • Search Google Scholar
    • Export Citation
  • 34.

    Goetze JP, Videbaek R, Boesgaard S, et al. Pro-brain natriuretic peptide as marker of cardiovascular or pulmonary causes of dyspnea in patients with terminal parenchymal lung disease. J Heart Lung Transplant 2004;23:8087.

    • Search Google Scholar
    • Export Citation
  • 35.

    Pruszczyk P. N-terminal pro-brain natriuretic peptide as an indicator of right ventricular dysfunction. J Card Fail 2005;11:S65S69.

  • 36.

    Pruszczyk P, Kostrubiec M, Bochowicz A, et al. N-terminal probrain natriuretic peptide in patients with acute pulmonary embolism. Eur Respir J 2003;22:649653.

    • Search Google Scholar
    • Export Citation
  • 37.

    Fijalkowska A, Kurzyna M, Torbicki A, et al. Serum N-terminal brain natriuretic peptide as a prognostic parameter in patients with pulmonary hypertension. Chest 2006;129:13131321.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 409 0 0
Full Text Views 1398 1103 26
PDF Downloads 548 282 26
Advertisement