Mitral valve disease is caused by degeneration of the mitral valve and is the most common acquired heart disease in dogs. It induces volume overload, which results in left ventricular eccentric hypertrophy and left atrial dilatation.1 Dogs with severe MVD can develop congestive heart failure, and the prognosis for such dogs is poor.2
The circulating hormones ANP and B-type natriuretic peptide are primarily synthesized by the heart and released into the blood in response to atrial and ventricular pressures.3,4 Cardiac troponin is a cardiac-specific protein, which contains 3 subunits (T, C, and I) that are released into the blood in response to cardiomyocyte damage.5,6 Results of previous studies7–10 indicate that those cardiac biomarkers are clinically useful for assessment of heart disease in dogs and cats. However, differential regulation of the secretory mechanisms that control the release of those cardiac biomarkers has been described.11,12 For example, in humans, circulating ANP concentration increases in response to atrial overload, whereas circulating
B-type natriuretic peptide concentration increases in response to ventricular overload.12 In dogs with experimentally induced ischemic myocardial injury, plasma cardiac troponin-T concentrations increased within 24 hours after myocardial infarction and remained increased for 7 days, whereas plasma ANP concentrations remained within reference limits.13
Cardiac troponin-I assays are available for clinical use10,14–16; however, the cTnI detection and reference limits vary among the assays. One assaya in particular has a low cTnI detection limit.17,18 In studies10,18,19 that used that assay to evaluate serum cTnI concentrations of dogs, dogs with MVD had a significantly greater serum cTnI concentration than healthy control dogs, and the cTnI concentration tended to increase as MVD progressed. To our knowledge, differences in the regulation of ANP and cTnI have not been investigated in dogs with naturally occurring MVD. The purpose of the study reported here was to evaluate and compare the clinical usefulness of plasma ANP and cTnI concentrations for assessment of disease severity in dogs with naturally occurring MVD.
Materials and Methods
Animals
The study was conducted in accordance with the experimental animal guidelines of the Japanese Ministry of Education, Culture, Sports, Science and Technology. It was a prospective multicenter study involving privately owned dogs that were examined from July 2015 to March 2017. All dogs were owned by veterinary students or clients. Owner consent was obtained for each dog prior to study enrollment. Each dog underwent a physical examination, blood sample collection, and echocardiographic and thoracic radiographic assessments, all of which were performed in a quiet examination room without the dog sedated. Information regarding previous treatment for cardiovascular disease was obtained from the owner of each dog.
To be included in the study, dogs with MVD had to be ≥ 6 years old and have a confirmed systolic heart murmur with a grade ≥ 3/6 that was auscultated over the mitral valve area. Mitral valve disease was defined as the presence of radiographic or echocardiographic evidence of characteristic valvular lesions on the mitral valve apparatus and mitral valve regurgitation on color Doppler echocardiography. Dogs with MVD were assigned to 1 of 3 disease stages on the basis of guidelines established by the American College of Veterinary Internal Medicine.1,20,21 Briefly, dogs without clinical signs of cardiac disease and an LA:Ao < 1.7 or LVIDDn < 1.7 cm/kg0.294 were assigned to stage B1. Dogs without clinical signs of cardiac disease and an LA:Ao ≥ 1.7 and LVIDDn ≥ 1.7 cm/kg0.294 were assigned to stage B2. Dogs with a history or clinical signs of congestive heart failure were assigned to stage C. Congestive heart failure, or cardiogenic pulmonary edema, was diagnosed on the basis of the presence of dyspnea and radiographic evidence of pulmonary infiltrates with an interstitial or alveolar pattern in the caudodorsal lung fields. For practical reasons, dogs that were prescribed cardiovascular medications were included in the study; however, dogs with MVD and 1 or more concurrent systemic illnesses such as pulmonary, endocrine, renal, hepatic, or inflammatory disease or malignances were excluded from the study. Dogs with no cardiac abnormalities as determined on the basis of results of physical, echocardiographic, and thoracic radiographic examinations and a serum biochemical profile were enrolled in the study as control dogs.
Thoracic radiography and echocardiography
For each dog, the cardiac silhouette was measured on thoracic radiographs by use of the vertebral heart scale as described.22,23 All transthoracic echocardiographic examinations were performed by experienced echocardiographers and use of ultrasonographic units with 7.5- to 12-MHz probes. The LA:Ao was measured on the right parasternal short-axis view by use of the 2-D method. M-mode echocardiography was also performed on the right parasternal short-axis view. Fractional shortening was calculated as (LVIDd - left ventricular internal dimension at end systole)/LVIDd × 100. The LVIDDn was calculated as LVIDd (measured in cm)/body weight (measured in kg)0.294 as described.20,24 Continuous Doppler echocardiography was used to measure the peak systolic mitral valve regurgitant flow velocity with the sample volume positioned at the tip of the mitral valve leaflets in the left parasternal long-axis view; the mean for 3 cardiac cycles was calculated and used for analysis purposes.
Measurement of cardiac biomarker concentrations
From each dog during the initial examination, blood (approx 3 mL) was obtained by jugular venipuncture and immediately placed into 2 blood collection tubes. One of the tubes contained aprotinin (aprotinin tube) as an anticoagulant, and the other contained heparin (heparin tube) as an anticoagulant. The blood tubes were centrifuged at 1,500 × g and 4°C for 10 minutes. Plasma was harvested from each sample, placed in a cryovial, and frozen within 1 hour after blood sample collection. All plasma samples were stored frozen at −20°C and shipped in a frozen state to a commercial laboratoryb for measurement of ANP and cTnI concentrations within 3 days after blood sample collection.
The ANP concentration was determined for plasma samples harvested from blood samples placed in aprotinin tubes, and the cTnI concentration was determined for plasma samples harvested from blood samples placed in heparin tubes. All assays were performed in duplicate, and the lab technician who performed the assays was unaware of the history and diagnosis for each dog.
Plasma ANP concentration was measured by use of an automated chemiluminescence enzyme immunoassayc for human αANP (C-terminal ANP). The manufacturer-stated detection range for ANP was 5 to 2,000 pg/mL. Plasma samples with an ANP concentration below the assay detection limit were assigned a concentration of 5 pg/mL for statistical analysis.
Plasma cTnI concentration was measured by use of a chemiluminescent immunoassaya for human cTnI. The assay was a second-generation, 3-site, sandwich immunoassay with a manufacturer-stated detection range for cTnI of 0.006 to 50.0 ng/mL. Plasma samples with a cTnI concentration below the assay detection limit were assigned a concentration of 0.006 ng/mL for statistical analysis.
According to the manufacturer, the cTnI assay has been validated for use with either plasma or serum. To confirm the variation of cTnI concentration owing to differences in anticoagulants, a blood sample (approx 6 mL) was collected from each of 2 healthy dogs and 5 dogs with heart disease and placed into 4 blood collection tubes including 1 tube without any additives (for serum), and 1 tube each with heparin, aprotinin, or EDTA as an anticoagulant (for plasma). The mean ± SD cTnI concentration for serum samples (0.179 ± 0.203 ng/mL) was similar to that for plasma samples harvested from EDTA-anticoagulated (0.165 ± 0.178 ng/mL), heparin-anticoagulated (0.161 ± 0.180 ng/mL), and aprotinin-anticoagulated (0.157 ± 0.166 ng/mL) blood samples.
For both the ANP and cTnI assays, the intra-assay and interassay coefficients of variation were determined for samples with known analyte concentrations that were representative of the low, middle, and high portions of the manufacturer-provided analyte detection range. Blood samples were repeatedly measured 5 times, and the coefficient of variation was calculated as follows; SD/mean value × 100.
To confirm dilutional parallelism of the assays, standard (human) solutions and plasma samples from dogs with heart failure were serially diluted with saline (0.9% NaCl) solution to achieve dilutions of 1 × 1, 2−1, 2−2, 2−3, 2−4, 2−5, 2−6, and 2−7. The plasma sample used for the ANP assay was obtained from a 12-year-old female Chihuahua with heart failure, in which the plasma ANP concentration was 623.3 pg/mL. The plasma sample used for the cTnI assay was obtained from a 16-year-old female Miniature Dachshund with a cardiac mass, in which the plasma cTnI concentration was 9.754 ng/mL.
Statistical analysis
All analyses were performed with statistical software.d,e Descriptive data were generated for each of the 4 study groups (stage B1, stage B2, stage C, and control). Data distributions for continuous variables were assessed for normality by means of the Kolmogorov-Smirnov test. None of the variables were normally distributed; therefore, results were reported as the median (IQR). The Kruskal-Wallis test was used to compare variables among the 4 study groups, with the Dunn test used for post hoc pairwise comparisons when necessary. Receiver operating characteristic curve analyses were used to assess cutoff values and predict the accuracy of plasma ANP and cTnI concentrations for detection of MVD severity. The Youden index was used to determine the optimal cutoff plasma concentrations of plasma ANP and cTnI to distinguish each MVD group. The ROC curves were compared by a nonparametric method as described.25 Values of P < 0.05 were considered significant for all analyses.
Results
Dogs
The study population consisted of 316 dogs with MVD and 40 healthy control dogs. The dogs with MVD included 115 Chihuahuas, 41 Cavalier King Charles Spaniels, 23 Toy Poodles, 22 Shih Tzus, 19 Maltese, 17 mixed-breed dogs, 14 Miniature Schnauzers, 13 Miniature Dachshunds, 12 Papillons, 12 Pomeranians, 4 Beagles, 4 Border Collies, 4 Yorkshire Terriers, 2 Boston Terriers, 2 Shetland Sheepdogs, 2 Spitzes, and 1 each of the following breeds: American Cocker Spaniel, Golden Retriever, Labrador Retriever, Miniature Pinscher, Pembroke Welsh Corgi, and Shiba Inu; the breed was not specified for 4 dogs with MVD. The control dogs included 9 Miniature Dachshunds, 6 Chihuahuas, 4 Toy Poodles, 3 mixed-breed dogs, 3 Pembroke Welsh Corgis, 3 Shih Tzus, 2 Miniature Pinschers, 2 Yorkshire Terriers, and 1 each of the following breeds: American Cocker Spaniel, Border Collie, Cavalier King Charles Spaniel, Maltese, Papillon, Labrador Retriever, and Shiba Inu; the breed was not specified for 1 control dog.
Of the 316 dogs with MVD, 142 (45%) were classified in the stage B1 group, 102 (32%) were classified in the stage B2 group, and 72 (23%) were classified in the stage C group. One hundred sixty-three (52%) and 153 (48%) dogs with MVD were and were not receiving medications for the treatment of cardiovascular disease, respectively, at the time of blood sample collection. Prescribed medications or medication types included angiotensin-converting enzyme inhibitors (n = 149 dogs), pimobendan (100), loop diuretics (59), calcium channel blockers (34), spironolactone (24), nitric oxide (10), β-adrenergic receptor blockers (8), beraprost sodium (4), sildenafil (4), cilostazol (2), digoxin (2), angiotensin-II receptor blockers (1), and thiazide (1); some dogs were receiving multiple medications or medication types. Additional descriptive data for each group were summarized (Table 1).
Select descriptive, radiographic, and echocardiographic variables for healthy control dogs (n = 40) and dogs with stage B1 (142), B2 (102), or C (72) MVD.
| Variable | Control | Stage B1 MVD | Stage B2 MVD | Stage C MVD |
|---|---|---|---|---|
| Age (y) | 9.5 (7.7–12.0) | 10.0 (8.0–12.0) | 10.0 (9.0–12.0) | 11.0 (9.1–12.2) |
| Sex | ||||
| Male | 17 | 72 | 58 | 39 |
| Female | 23 | 70 | 44 | 33 |
| Body weight (kg) | 4.8 (3.6–7.2) | 4.8 (3.5–7.5) | 3.8 (2.9–6.3)† | 3.6 (2.9–6.0)† |
| Heart rate (beats/min) | 120 (108–140) | 126 (l12–149) | 144 (129–168)* | 152 (140–178)*† |
| Vertebral heart score | 9.8 (9.3–10.1) | 10.4 (9.8–11.0)* | 11.2 (10.8–11.9)*† | 12.2 (11.2–13.7)*† |
| IVSd (mm) | 6.9 (5.6–8.0) | 6.0 (5.0–8.0) | 5.8 (5.0–6.6)* | 5.8 (4.9–7.0) |
| LVIDd (mm) | 19.4 (18.5–25.0) | 26.0 (22.0–30.0)* | 31.0 (28.2–34.7)*† | 32.7 (27.0–37.8)† |
| LVPWd (mm) | 7.0 (5.6–8.0) | 6.3 (5.6–7.2) | 5.6 (4.8–6.6)*† | 6.3 (5.1–7.1) |
| LVIDDn (cm/kg0.294) | 1.25 (1.06–1.60) | 1.61 (1.49–1.74) | 2.05 (1.93–2.24) | 2.10 (1.76–2.51) |
| Fractional shortening (%) | 40.7 (35.5–48.8) | 45.6 (39.2–53.8) | 52.6 (48.6–57.1)*† | 50.5 (45.0–57.8)† |
| LA:Ao | 1.2 (1.1–1.3) | 1.5 (1.3–1.6) | 2.1 (1.9–2.4) | 2.5 (2.0–3.0) |
| Mitral valve regurgitant flow (m/s) | — | 5.7 (5.2–6.1) | 5.4 (5.1–5.9)† | 5.0 (4.6–5.4)†‡ |
Values represent the number of dogs or median (IQR). Dogs with MVD were categorized on the basis of disease severity in accordance with guidelines established by the American College of Veterinary Internal Medicine. Dogs in stage B1 (n = 142) had no clinical signs of cardiac disease and an LA:Ao < 1.7 or LVIDDn < 1.7 cm/kg0.294. Dogs in stage B2 (n = 102) had no clinical signs of cardiac disease and an LA:Ao ≥ 1.7 or LVIDDn > 1.7 cm/kg0.294. Dogs in group C (n = 72) had a history of or clinical signs consistent with congestive heart failure, which was defined as dyspnea and radiographic evidence of pulmonary infiltrates with an interstitial or alveolar pattern in the caudodorsal lung fields.
Value differs significantly (P < 0.05) from that for the control group.
Value differs significantly (P < 0.05) from that for the stage B1 group.
Value differs significantly (P < 0.05) from that for the stage B2 group.
IVSd = Interventricular septal thickness at end diastole. LVPWd = Left ventricular posterior wall thickness at end diastole. — = Not applicable.
Select radiographic and echocardiographic indices for each group were also summarized (Table 1). The median vertebral heart score and LVIDd for dogs with MVD were significantly greater than the corresponding values for the control group. The median LVIDd for dogs in the stage B2 and C groups was also significantly greater than that for dogs in the stage B1 group. The median heart rate and fractional shortening for dogs in the stage B2 and C groups were significantly greater than the corresponding values for dogs in the control and stage B1 groups. For dogs with MVD, the mitral valve regurgitant flow velocity decreased significantly as disease severity increased.
Assay performance
For both the ANP and cTnI assays, the intra-assay and interassay coefficients of variation for samples with known analyte concentrations at the low, middle, and high portions of the detection range were summarized (Table 2). Results indicated that the consistency was acceptable for both assays when used to evaluate canine plasma samples. Dilutional parallelism was confirmed for both assays (Figure 1).
Counts per minute of absorbance for serially diluted canine plasma samples (black circles, with dilutions indicated) and human standards (white circles), as measured by use of an automated chemiluminescence enzyme immunoassayb for human αANP (C-terminal ANP; A) and a chemiluminescent immunoassayc for human cTnI (B). Samples were diluted with saline (0.9% NaCl) solution. A—The plasma sample was obtained from a 12-year-old female Chihuahua with heart failure, in which the plasma ANP concentration was 623.3 pg/mL. The ANP concentration changed concomitantly with serial dilution as follows: 270.9 pg/mL at 1 × 2−1, 127.6 pg/mL at 1 × 2−2, 65.3 pg/mL at 1 × 2−3, 34.3 pg/mL at 1 × 2−4, 40.3 pg/mL at 1 × 2−5, 11.2 pg/mL at 1 × 2−6, and 8.4 pg/mL at 1 × 2−7. B—The plasma sample was obtained from a 16-year-old female Miniature Dachshund with a cardiac mass, in which the plasma cTnI concentration was 9.754 ng/mL. The cTnI concentration changed concomitantly with serial dilution as follows: 4.936 ng/mL at 1 × 2−1, 2.469 ng/mL at 1 × 2−2, 1.256 ng/mL at 1 × 2−3, 0.565 ng/mL at 1 × 2−4, 0.295 ng/mL at 1 × 2−5, 0.130 ng/mL at 1 × 2−6, and 0.061 ng/mL at 1 × 2−7. Parallelism was evident between the human standard and canine plasma sample for each assay.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Counts per minute of absorbance for serially diluted canine plasma samples (black circles, with dilutions indicated) and human standards (white circles), as measured by use of an automated chemiluminescence enzyme immunoassayb for human αANP (C-terminal ANP; A) and a chemiluminescent immunoassayc for human cTnI (B). Samples were diluted with saline (0.9% NaCl) solution. A—The plasma sample was obtained from a 12-year-old female Chihuahua with heart failure, in which the plasma ANP concentration was 623.3 pg/mL. The ANP concentration changed concomitantly with serial dilution as follows: 270.9 pg/mL at 1 × 2−1, 127.6 pg/mL at 1 × 2−2, 65.3 pg/mL at 1 × 2−3, 34.3 pg/mL at 1 × 2−4, 40.3 pg/mL at 1 × 2−5, 11.2 pg/mL at 1 × 2−6, and 8.4 pg/mL at 1 × 2−7. B—The plasma sample was obtained from a 16-year-old female Miniature Dachshund with a cardiac mass, in which the plasma cTnI concentration was 9.754 ng/mL. The cTnI concentration changed concomitantly with serial dilution as follows: 4.936 ng/mL at 1 × 2−1, 2.469 ng/mL at 1 × 2−2, 1.256 ng/mL at 1 × 2−3, 0.565 ng/mL at 1 × 2−4, 0.295 ng/mL at 1 × 2−5, 0.130 ng/mL at 1 × 2−6, and 0.061 ng/mL at 1 × 2−7. Parallelism was evident between the human standard and canine plasma sample for each assay.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Counts per minute of absorbance for serially diluted canine plasma samples (black circles, with dilutions indicated) and human standards (white circles), as measured by use of an automated chemiluminescence enzyme immunoassayb for human αANP (C-terminal ANP; A) and a chemiluminescent immunoassayc for human cTnI (B). Samples were diluted with saline (0.9% NaCl) solution. A—The plasma sample was obtained from a 12-year-old female Chihuahua with heart failure, in which the plasma ANP concentration was 623.3 pg/mL. The ANP concentration changed concomitantly with serial dilution as follows: 270.9 pg/mL at 1 × 2−1, 127.6 pg/mL at 1 × 2−2, 65.3 pg/mL at 1 × 2−3, 34.3 pg/mL at 1 × 2−4, 40.3 pg/mL at 1 × 2−5, 11.2 pg/mL at 1 × 2−6, and 8.4 pg/mL at 1 × 2−7. B—The plasma sample was obtained from a 16-year-old female Miniature Dachshund with a cardiac mass, in which the plasma cTnI concentration was 9.754 ng/mL. The cTnI concentration changed concomitantly with serial dilution as follows: 4.936 ng/mL at 1 × 2−1, 2.469 ng/mL at 1 × 2−2, 1.256 ng/mL at 1 × 2−3, 0.565 ng/mL at 1 × 2−4, 0.295 ng/mL at 1 × 2−5, 0.130 ng/mL at 1 × 2−6, and 0.061 ng/mL at 1 × 2−7. Parallelism was evident between the human standard and canine plasma sample for each assay.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Intra-assay and interassay coefficients of variation for an automated chemiluminescence enzyme immunoassayc for human αANP (C-terminal ANP) and a chemiluminescent immunoassaya for human cTnI when used to assess ANP and cTnI concentrations, respectively, in canine plasma.
| Portion of assay detection range | ||||
|---|---|---|---|---|
| Assay | Variable | Low | Middle | High |
| ANP | Mean measured ANP concentration for plasma samples used to calculate intra-assay coefficient of variation (pg/mL) | 59.9 | 148.4 | 514.0 |
| Intra-assay coefficient of variation (%) | 3.0 | 3.4 | 7.0 | |
| Mean measured ANP concentration for plasma samples used to calculate interassay coefficient of variation (pg/mL) | 61.2 | 153.0 | 511.1 | |
| Interassay coefficient of variation (%) | 3.1 | 2.6 | 3.1 | |
| cTnI | Mean measured cTnI concentration for plasma samples used to calculate intra-assay coefficient of variation (ng/mL) | 0.008 | 0.417 | 47.510 |
| Intra-assay coefficient of variation (%) | 10.7 | 2.3 | 2.1 | |
| Mean measured cTnI concentration for plasma samples used to calculate interassay coefficient of variation (ng/mL) | 0.008 | 0.424 | 48.026 | |
| Interassay coefficient of variation (%) | 7.7 | 2.9 | 2.0 | |
The manufacturer-provided analyte detection range was 5 to 2,000 pg/mL for the ANP assay and 0.006 to 50 ng/mL for the cTnI assay.
Plasma ANP and cTnI concentrations
The median plasma ANP concentrations for the stage C (436.7 pg/mL; IQR, 219.9 to 754.3 pg/mL), stage B2 (274.2 pg/mL; IQR, 151.1 to 542.6 pg/mL), and stage B1 (105.3 pg/mL; IQR, 65.6 to 177.9 pg/mL) groups were significantly greater than the median plasma ANP concentration for the control group (61.9 pg/mL; IQR, 47.6 to 79.2 pg/mL). The median plasma ANP concentrations for the stage B2 and C groups were also significantly greater than the median plasma ANP concentration for the stage B1 group (Figure 2).
Box-and-whisker plots of plasma ANP (A) and cTnI (B) concentration for 40 healthy control dogs and 316 dogs with MVD that were categorized into 3 groups on the basis of disease severity and guidelines established by the American College of Veterinary Internal Medicine. Dogs in stage B1 (n = 142) had no clinical signs of cardiac disease and an LA:Ao < 1.7 or LVlDDn < 1.7 cm/kg0.294. Dogs in stage B2 (n = 102) had no clinical signs of cardiac disease and LA:Ao ≥ 1.7 or LVIDDn ≥ 1.7 cm/kg0.294. Dogs in group C (n = 72) had a history of or clinical signs consistent with congestive heart failure, which was defined as dyspnea and radiographic evidence of pulmonary infiltrates with an interstitial or alveolar pattern in the caudodorsal lung fields. For each plot, the small square represents the mean, the horizontal line within each box represents the median, the lower and upper borders of the box represent the 25th and 75th percentiles, respectively, and the whiskers delimit the 10th and 90th percentiles. *Median value differs significantly (P < 0.05) from that for the control group. †Median value differs significantly (P < 0.01) from that for the stage B1 group. ‡Median value differs significantly (P < 0.001) from that for the stage B2 group.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Box-and-whisker plots of plasma ANP (A) and cTnI (B) concentration for 40 healthy control dogs and 316 dogs with MVD that were categorized into 3 groups on the basis of disease severity and guidelines established by the American College of Veterinary Internal Medicine. Dogs in stage B1 (n = 142) had no clinical signs of cardiac disease and an LA:Ao < 1.7 or LVlDDn < 1.7 cm/kg0.294. Dogs in stage B2 (n = 102) had no clinical signs of cardiac disease and LA:Ao ≥ 1.7 or LVIDDn ≥ 1.7 cm/kg0.294. Dogs in group C (n = 72) had a history of or clinical signs consistent with congestive heart failure, which was defined as dyspnea and radiographic evidence of pulmonary infiltrates with an interstitial or alveolar pattern in the caudodorsal lung fields. For each plot, the small square represents the mean, the horizontal line within each box represents the median, the lower and upper borders of the box represent the 25th and 75th percentiles, respectively, and the whiskers delimit the 10th and 90th percentiles. *Median value differs significantly (P < 0.05) from that for the control group. †Median value differs significantly (P < 0.01) from that for the stage B1 group. ‡Median value differs significantly (P < 0.001) from that for the stage B2 group.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Box-and-whisker plots of plasma ANP (A) and cTnI (B) concentration for 40 healthy control dogs and 316 dogs with MVD that were categorized into 3 groups on the basis of disease severity and guidelines established by the American College of Veterinary Internal Medicine. Dogs in stage B1 (n = 142) had no clinical signs of cardiac disease and an LA:Ao < 1.7 or LVlDDn < 1.7 cm/kg0.294. Dogs in stage B2 (n = 102) had no clinical signs of cardiac disease and LA:Ao ≥ 1.7 or LVIDDn ≥ 1.7 cm/kg0.294. Dogs in group C (n = 72) had a history of or clinical signs consistent with congestive heart failure, which was defined as dyspnea and radiographic evidence of pulmonary infiltrates with an interstitial or alveolar pattern in the caudodorsal lung fields. For each plot, the small square represents the mean, the horizontal line within each box represents the median, the lower and upper borders of the box represent the 25th and 75th percentiles, respectively, and the whiskers delimit the 10th and 90th percentiles. *Median value differs significantly (P < 0.05) from that for the control group. †Median value differs significantly (P < 0.01) from that for the stage B1 group. ‡Median value differs significantly (P < 0.001) from that for the stage B2 group.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
The median plasma cTnI concentration for the stage C group (0.249 ng/mL; IQR, 0.168 to 0.784 ng/mL) was significantly greater than that for each of the other 3 study groups. The median plasma cTnI concentration for the stage B2 group (0.098 ng/mL; IQR, 0.058 to 0.182 ng/mL) was significantly greater than the median plasma cTnI concentrations for the stage B1 (0.068 ng/mL; IQR, 0.045 to 0.122 ng/mL) and control (0.058 ng/mL; IQR, 0.038 to 0.088 ng/mL) groups. The median plasma cTnI did not differ significantly between the stage B1 and control groups (Figure 2).
The assay test characteristics as determined by ROC analyses were summarized (Table 3). The plasma ANP concentrations identified as the optimal cutoffs were 112.9, 184.2, and 277.8 pg/mL for detection of dogs with stage B1 or worse, stage B2 or worse, and stage C MVD, respectively. The plasma cTnI concentrations identified as the optimal cutoffs were 0.089, 0.139, and 0.163 ng/mL for detection of dogs with stage B1 or worse, stage B2 or worse, and stage C MVD, respectively. The AUCs for the ANP concentration cutoffs for detection of dogs with stage B1 or worse (P < 0.001) and stage B2 or worse MVD (P = 0.001) were significantly greater than the AUCs for the corresponding cTnI concentration cutoffs. However, the AUCs for the ANP and cTnI concentration cutoffs for detection of dogs with stage C MVD were similar. The ROC curves likewise indicated that plasma ANP concentration was better than plasma cTnI concentration for detection of dogs in the earliest stages (B1 and B2) of MVD, but the discerning ability of the 2 biomarkers was similar for dogs with advanced (stage C) MVD (Figure 3).
Receiver operating characteristic curve analyses for the use of plasma ANP and cTnI concentrations as determined by the assays described in Table 2 to identify dogs with stage B1, B2, or C MVD.
| MVD stage | ||||
|---|---|---|---|---|
| Assay | Test characteristic | B1 | B2 | C |
| ANP | Cutoff concentration (pg/mL) | 112.9 | 184.2 | 277.8 |
| AUC | 0.85 (0.80–0.90)* | 0.86 (0.82–0.90)* | 0.79 (0.73–0.86) | |
| Sensitivity (%) | 70.1 (64.6–75.3) | 72.1 (64.6–78.8) | 70.2 (57.7–80.7) | |
| Specificity (%) | 94.3 (80.8–99.3) | 82.7 (76.2–88.1) | 78.6 (73.1–83.3) | |
| Positive predictive value (%) | 99.0 (97.0–99.7) | 80.4 (75.1–84.9) | 45.2 (38.6–50.8) | |
| Negative predictive value (%) | 26.8 (23.3–28.0) | 75.1 (70.9–78.8) | 91.3 (88.3–93.8) | |
| Positive likelihood ratio | 12.2 (3.8–44.2) | 4.2 (3.1–5.7) | 3.3 (2.5–4.1) | |
| Negative likelihood ratio | 0.32 (0.30–0.39) | 0.34 (0.27–0.42) | 0.38 (0.26–0.53) | |
| cTnI | Cutoff concentration (ng/mL) | 0.089 | 0.139 | 0.163 |
| AUC | 0.69 (0.61–0.77) | 0.74 (0.69–0.79) | 0.84 (0.78–0.89) | |
| Sensitivity (%) | 54.5 (48.8–60.1) | 54.9 (47.2–62.5) | 76.4 (64.9–85.6) | |
| Specificity (%) | 78.1 (62.4–89.4) | 83.0 (76.7–88.1) | 81.6 (76.6–86.0) | |
| Positive predictive value (%) | 95.0 (91.8–97.2) | 78.5 (71.9–84.2) | 66.3 (58.7–72.4) | |
| Negative predictive value (%) | 18.3 (15.0–20.6) | 66.7 (62.3–69.6) | 93.8 (91.4–95.6) | |
| Positive likelihood ratio | 2.5 (1.5–4.6) | 3.8 (2.7–5.6) | 7.7 (5.6–10.3) | |
| Negative likelihood ratio | 0.58 (0.50–0.74) | 0.53 (0.46–0.61) | 0.26 (0.18–0.37) | |
Values in parentheses represent the 95% confidence interval. Within each column, results reflect test characteristics for detection of dogs with MVD at the given stage or worse.
Value differs significantly (P < 0.05) from the corresponding value for the cTnI assay.
Receiver operating characteristic curves for the use of plasma ANP (black circles) and cTnI (white circles) concentrations to distinguish between the dogs of Figure 2 with and without stage B1 or worse (A), stage B2 or worse (B), or stage C (C) MVD. Notice that plasma ANP concentration had better discerning ability than plasma cTnI concentration for identification of dogs in the earliest (subclinical) stages (B1 and B2) of MVD. See Figure 2 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Receiver operating characteristic curves for the use of plasma ANP (black circles) and cTnI (white circles) concentrations to distinguish between the dogs of Figure 2 with and without stage B1 or worse (A), stage B2 or worse (B), or stage C (C) MVD. Notice that plasma ANP concentration had better discerning ability than plasma cTnI concentration for identification of dogs in the earliest (subclinical) stages (B1 and B2) of MVD. See Figure 2 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Receiver operating characteristic curves for the use of plasma ANP (black circles) and cTnI (white circles) concentrations to distinguish between the dogs of Figure 2 with and without stage B1 or worse (A), stage B2 or worse (B), or stage C (C) MVD. Notice that plasma ANP concentration had better discerning ability than plasma cTnI concentration for identification of dogs in the earliest (subclinical) stages (B1 and B2) of MVD. See Figure 2 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 256, 3; 10.2460/javma.256.3.340
Discussion
In the present study, ANP and cTnI concentrations in canine plasma were measured by means of commercial assays validated and marketed for use in human medicine that used anti-human ANP and cTnI antibodies. The amino acid sequences of ANP and cTnI are highly homologous between humans and dogs.17,26,27 Parallelism between canine plasma and both human ANP and cTnI standards was confirmed, and the intra-assay and interassay coefficients of variation calculated for the ANP and cTnI assays were sufficient to indicate that the assays were appropriate for measurement of ANP and cTnI concentrations in canine plasma. Collectively, these results indicated that anti-human ANP and cTnI antibodies have affinity for canine ANP and cTnI, respectively.
The median plasma ANP concentration was 61.9 pg/mL (IQR, 47.6 to 79.2 pg/mL) for the healthy control dogs of the present study. For healthy dogs of other studies, the reported mean ± SD plasma ANP concentration ranges from 16.4 ± 7.8 pg/mL (n = 37 dogs)8 to 56.9 ± 17.4 pg/mL (10 dogs).28 Variations in the plasma ANP concentration among healthy dogs of those studies8,28 may reflect differences in the study population or methodology used to measure ANP concentration. An automated chemiluminescence enzyme immunoassay was used to measure plasma ANP concentration in the present study, whereas a radioimmunoassay was used to measure the plasma ANP concentration in those other studies.8,28
The median plasma cTnI concentration was 0.058 ng/mL (IQR, 0.038 to 0.088 ng/mL) for the control dogs of the present study. By contrast, the median serum cTnI concentration was 0.017 ng/mL (range, 0.006 to 0.136 ng/mL) for 30 healthy dogs of another study.19 The addition of heparin to canine and human serum samples decreases the measured cTnI concentration.10,29 The reason that the plasma cTnI concentration for the healthy control dogs of the present study was 3.4 times the serum cTnI concentration for the healthy dogs of that other study19 was unknown. Further research is necessary to establish reference intervals for ANP and cTnI concentrations in dogs.
In dogs, plasma ANP and N-terminal proANP concentrations increase with the severity of MVD.8,30,31 Results of the present study were consistent with those findings. Further, in the present study, the median plasma ANP concentrations for dogs with subclinical MVD (stages B1 and B2) were significantly greater than the median plasma ANP concentration for the control dogs. However, in a study31 of Cavalier King Charles Spaniels, the mean plasma ANP concentration for dogs with subclinical MVD did not differ significantly from that for healthy control dogs. The apparently conflicting results between the present study and that study31 most likely reflect differences in the study populations (particularly breed distribution) and criteria used to classify severity of MVD. The left ventricular and left atrial dimensions did not differ significantly between the dogs with subclinical MVD and control dogs of that other study,31 whereas the left ventricular and left atrial dimensions for dogs with subclinical MVD were significantly greater than those for the control dogs of the present study. Left atrial congestion is an important regulator of ANP concentration.3,30,32,33 It is possible that MVD-induced left atrial enlargement causes the plasma ANP concentration to increase even before clinical signs of the disease become evident.
The plasma cTnI concentration is abnormally increased in dogs with moderate and severe MVD,19,34 and cTnI concentration is negatively associated with prognosis.35 The same chemiluminescent immunoassay for human cTnI as that used in the present study has been used to measure the serum cTnI concentration for dogs of other studies,10,18,19 and results of those studies indicate that the serum cTnI concentration is greater in dogs with MVD than in healthy control dogs. However, those studies10,18,19 did not analyze the effect of disease severity on serum cTnI concentration. A larger number of dogs were evaluated in the present study than in those studies,10,18,19 and the dogs with MVD were categorized on the basis of disease severity. There was a sufficient number of dogs in each category to allow plasma cTnI concentration to be compared among categories. Although the plasma cTnI concentration increased with MVD severity among dogs with MVD in the present study, it did not differ significantly between dogs in the earliest stage (B1) of the disease and control dogs. Moreover, there was substantial overlap in the plasma cTnI concentration ranges for the control dogs and dogs with subclinical MVD (stages B1 and B2). Cardiac troponin is released into the blood subsequent to cardiomyocyte damage,5,6,36 and left ventricular interstitial cells in dogs with MVD are in a proapoptotic state but do not undergo apoptosis.37 Therefore, we speculated that dogs with subclinical MVD have only slight myocardial injury.
Although results of several studies8,15,35,38 indicate that circulating ANP and cTnI concentrations increase as MVD severity increases, data comparing the diagnostic usefulness of plasma ANP and cTnI concentrations for dogs with MVD were lacking prior to the present study. In the present study, the AUCs for ANP concentration for dogs with MVD in stages B1 and B2 were significantly greater than the corresponding AUCs for cTnI concentration. Therefore, plasma ANP concentration appeared to be better than plasma cTnI concentration for detection of dogs with subclinical MVD. Additionally, a plasma ANP concentration > 112.9 pg/mL was highly specific (94.3%) for dogs with stage B1 MVD, and a plasma ANP concentration > 184.2 pg/mL had moderate sensitivity (72.1%) and specificity (82.7%) for detection of dogs with stage B2 MVD. Those results indicated that a plasma ANP concentration of ≤ 112.9 pg/mL was an excellent cutoff for excluding MVD as a diagnosis, whereas a plasma ANP concentration > 184.2 pg/mL may be a useful cutoff for identifying dogs with at least stage B2 MVD. The AUC for ANP concentration (0.79) was similar to the AUC for cTnI concentration (0.84) for dogs with stage C MVD. Thus, a plasma ANP concentration > 277.8 pg/mL and cTnI concentration > 0.163 ng/mL appeared to be equally useful for identification of dogs with stage C MVD.
The present study had several limitations. Plasma cTnI concentrations can increase after any type of myocardial damage, such as that induced by systemic inflammatory disease, trauma, chronic kidney disease, and cardiac hemangiosarcoma,14,16,39 and is not specific for MVD. Also, dogs that were receiving cardiovascular medications for the treatment of MVD were not excluded from the study, and the medications might have affected the measured plasma ANP and cTnI concentrations such that the concentrations did not accurately reflect the true nonmedicated disease state. In healthy dogs, the distribution of natriuretic peptide varies by breed; specifically, the proANP 31–67 concentration is greater in German Shepherd Dogs and Cavalier King Charles Spaniels than in Doberman Pinschers.40 Dogs of several breeds were enrolled in the present study. Further research is necessary to define the relationship between plasma ANP concentration and heart disease for specific breeds. Finally, the present study was cross-sectional in nature, and the results should be interpreted with caution. A longitudinal study is necessary to investigate whether ANP and cTnI concentrations are associated with the risk of cardiogenic pulmonary edema.
To our knowledge, the present study was the first to compare plasma ANP and cTnI concentrations among healthy control dogs and dogs in various stages of MVD. Results indicated that plasma ANP concentration, but not plasma cTnI concentration, might be useful for detecting dogs with subclinical MVD. Plasma ANP concentration and plasma cTnI concentration both had a high sensitivity for detection of dogs with past or current congestive heart failure (stage C MVD). Although plasma ANP and cTnI concentrations can be used to support echocardiographic findings, they should not be used independently to diagnose MVD. However, plasma ANP and cTnI concentrations can be used to assess the severity of MVD in dogs.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
ABBREVIATIONS
| ANP | Atrial natriuretic peptide |
| AUC | Area under the receiver operating characteristic curve |
| cTnI | Cardiac troponin-I |
| IQR | Interquartile (25th to 75th percentile) range |
| LA:Ao | Left atrium-to-aortic root ratio |
| LVIDd | Left ventricular internal diameter at end diastole |
| LVIDDn | Left ventricular internal diameter in diastole normalized to body weight |
| MVD | Mitral valve disease |
| ROC | Receiver operating characteristic |
Footnotes
ADVIA Centaur CP TnI-ultra, Siemens Healthineers Japan, Tokyo, Japan.
FUJIFILM VET Systems Co Ltd, Tokyo, Japan.
CL-JACK NX, Kyowa Medex, Tokyo, Japan.
Statemate III, version 3.16, Avice Inc, Tokyo, Japan.
BellCurve for Excel, Social Survey Research Information Co Ltd, Tokyo, Japan.
References
1. Atkins C, Bonagura J, Ettinger S, et al. Guidelines for the diagnosis and treatment of canine chronic valvular heart disease. J Vet Intern Med 2009;23:1142–1150.
2. Borgarelli M, Savarino P, Crosara S, et al. Survival characteristics and prognostic variables of dogs with mitral regurgitation attributable to myxomatous valve disease. J Vet Intern Med 2008;22:120–128.
3. Hori Y, Ukai Y, Uechi M, et al. Relationships between velocities of pulmonary venous flow and plasma concentrations of atrial natriuretic peptide in healthy dogs. Am J Vet Res 2008;69:465–470.
4. Ruskoaho H. Cardiac hormones as diagnostic tools in heart failure. Endocr Rev 2003;24:341–356.
5. Remppis A, Ehlermann P, Giannitsis E, et al. Cardiac troponin T levels at 96 hours reflect myocardial infarct size: a pathoanatomical study. Cardiology 2000;93:249–253.
6. Voss EM, Sharkey SW, Gernert AE, et al. Human and canine cardiac troponin T and creatine kinase-MB distribution in normal and diseased myocardium. Infarct sizing using serum profiles. Arch Pathol Lab Med 1995;119:799–806.
7. Hori Y, Yamano S, Iwanaga K, et al. Evaluation of plasma C-terminal atrial natriuretic peptide in healthy cats and cats with heart disease. J Vet Intern Med 2008;22:135–139.
8. Hori Y, Yamano S, Kanai K, et al. Clinical implications of measurement of plasma atrial natriuretic peptide concentration in dogs with spontaneous heart disease. J Am Vet Med Assoc 2011;239:1077–1083.
9. Langhorn R, Tarnow I, Willesen JL, et al. Cardiac troponin I and T as prognostic markers in cats with hypertrophic cardiomyopathy. J Vet Intern Med 2014;28:1485–1491.
10. Winter RL, Saunders AB, Gordon SG, et al. Analytical validation and clinical evaluation of a commercially available high-sensitivity immunoassay for the measurement of troponin I in humans for use in dogs. J Vet Cardiol 2014;16:81–89.
11. Oosterhof T, Tulevski II, Vliegen HW, et al. Effects of volume and/or pressure overload secondary to congenital heart disease (tetralogy of fallot or pulmonary stenosis) on right ventricular function using cardiovascular magnetic resonance and B-type natriuretic peptide levels. Am J Cardiol 2006;97:1051–1055.
12. Yoshimura M, Yasue H, Okumura K, et al. Different secretion patterns of atrial natriuretic peptide and brain natriuretic peptide in patients with congestive heart failure. Circulation 1993;87:464–469.
13. Hori Y, Ohshima N, Chikazawa S, et al. Myocardial injury-related changes in plasma NT-proBNP and ANP concentrations in a canine model of ischemic myocardial injury. Vet J 2012;191:46–51.
14. Chun R, Kellihan HB, Henik RA, et al. Comparison of plasma cardiac troponin I concentrations among dogs with cardiac hemangiosarcoma, noncardiac hemangiosarcoma, other neoplasms, and pericardial effusion of nonhemangiosarcoma origin. J Am Vet Med Assoc 2010;237:806–811.
15. Falk T, Ljungvall I, Zois NE, et al. Cardiac troponin-I concentration, myocardial arteriosclerosis, and fibrosis in dogs with congestive heart failure because of myxomatous mitral valve disease. J Vet Intern Med 2013;27:500–506.
16. Porciello F, Rishniw M, Herndon WE, et al. Cardiac troponin I is elevated in dogs and cats with azotaemia renal failure and in dogs with non-cardiac systemic disease. Aust Vet J 2008;86:390–394.
17. Langhorn R, Willesen JL, Tarnow I, et al. Evaluation of a high-sensitivity assay for measurement of canine and feline serum cardiac troponin I. Vet Clin Pathol 2013;42:490–498.
18. Winter RL, Saunders AB, Gordon SG, et al. Biologic variability of cardiac troponin I in healthy dogs and dogs with different stages of myxomatous mitral valve disease using standard and high-sensitivity immunoassays. Vet Clin Pathol 2017;46:299–307.
19. Polizopoulou ZS, Koutinas CK, Dasopoulou A, et al. Serial analysis of serum cardiac troponin I changes and correlation with clinical findings in 46 dogs with mitral valve disease. Vet Clin Pathol 2014;43:218–225.
20. Boswood A, Haggstrom J, Gordon SG, et al. Effect of pimobendan in dogs with preclinical myxomatous mitral valve disease and cardiomegaly: the EPIC study—a randomized clinical trial. J Vet Intern Med 2016;30:1765–1779.
21. Gordon SG, Saunders AB, Wesselowski SR. Asymptomatic canine degenerative valve disease: current and future therapies. Vet Clin North Am Small Anim Pract 2017;47:955–975.
22. Buchanan JW, Bücheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc 1995;206:194–199.
23. Hamlin RL. Analysis of the cardiac silhouette in dorsoventral radiographs from dogs with heart disease. J Am Vet Med Assoc 1968;153:1446–1460.
24. Cornell CC, Kittleson MD, Della Torre P, et al. Allometric scaling of M-mode cardiac measurements in normal adult dogs. J Vet Intern Med 2004;18:311–321.
25. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988;44:837–845.
26. Biondo AW, Liu ZL, Wiedmeyer CE, et al. Genomic sequence and cardiac expression of atrial natriuretic peptide in cats. Am J Vet Res 2002;63:236–240.
27. Rishniw M, Barr SC, Simpson KW, et al. Cloning and sequencing of the canine and feline cardiac troponin I genes. Am J Vet Res 2004;65:53–58.
28. Greco DS, Biller B, Van Liew CH. Measurement of plasma atrial natriuretic peptide as an indicator of prognosis in dogs with cardiac disease. Can Vet J 2003;44:293–297.
29. Katrukha A, Bereznikova A, Filatov V, et al. Biochemical factors influencing measurement of cardiac troponin I in serum. Clin Chem Lab Med 1999;37:1091–1095.
30. Asano K, Masuda K, Okumura M, et al. Plasma atrial and brain natriuretic peptide levels in dogs with congestive heart failure. J Vet Med Sci 1999;61:523–529.
31. Häggström J, Hansson K, Kvart C, et al. Relationship between different natriuretic peptides and severity of naturally acquired mitral regurgitation in dogs with chronic myxomatous valve disease. J Vet Cardiol 2000;2:7–16.
32. Lang CC, Choy AM, Turner K, et al. The effect of intravenous saline loading on plasma levels of brain natriuretic peptide in man. J Hypertens 1993;11:737–741.
33. Wambach G, Koch J. BNP plasma levels during acute volume expansion and chronic sodium loading in normal men. Clin Exp Hypertens 1995;17:619–629.
34. Ljungvall I, Höglund K, Tidholm A, et al. Cardiac troponin I is associated with severity of myxomatous mitral valve disease, age, and C-reactive protein in dogs. J Vet Intern Med 2010;24:153–159.
35. Hezzell MJ, Boswood A, Chang YM, et al. The combined prognostic potential of serum high-sensitivity cardiac troponin I and N-terminal pro-B-type natriuretic peptide concentrations in dogs with degenerative mitral valve disease. J Vet Intern Med 2012;26:302–311.
36. O'Brien PJ, Dameron GW, Beck ML, et al. Differential reactivity of cardiac and skeletal muscle from various species in two generations of cardiac troponin-T immunoassays. Res Vet Sci 1998;65:135–137.
37. Surachetpong S, Jiranantasak T, Rungsipipat A, et al. Apoptosis and abundance of Bcl-2 family and transforming growth factor β1 signaling proteins in canine myxomatous mitral valves. J Vet Cardiol 2013;15:171–180.
38. Fonfara S, Loureiro J, Swift S, et al. Cardiac troponin I as a marker for severity and prognosis of cardiac disease in dogs. Vet J 2010;184:334–339.
39. Hamacher L, Dorfelt R, Muller M, et al. Serum cardiac troponin I concentrations in dogs with systemic inflammatory response syndrome. J Vet Intern Med 2015;29:164–170.
40. Sjöstrand K, Wess G, Ljungvall I, et al. Breed differences in natriuretic peptides in healthy dogs. J Vet Intern Med 2014;28:451–457.
