- Email: [email protected]
Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease
Journal of Veterinary Cardiology (2016)
-, -e-
www.elsevier.com/locate/jvc
Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease Randolph L. Winter, DVM a,*, Ashley B. Saunders, DVM a, Sonya G. Gordon, DVM, DVSc a, Jesse S. Buch, PhD b, Matthew W. Miller, DVM, MS a a
Department of Small Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX, USA b Immunoassay R&D, IDEXX Laboratories, Westbrook, ME, USA Received 26 June 2016; received in revised form 11 November 2016; accepted 21 November 2016
KEYWORDS Canine; Cardiac biomarker; Mitral regurgitation; Prospective
Abstract Introduction: To determine the biologic variability of N-terminal probrain natriuretic peptide (NTproBNP) in healthy dogs and dogs with various stages of myxomatous mitral valve disease (MMVD). Animals: Thirty-eight privately owned dogs: 28 with MMVD and 10 healthy controls. Materials and methods: Prospective clinical study with comprehensive evaluation used to group dogs as healthy or into three stages of MMVD based on current guidelines. NTproBNP was measured hourly, daily, and weekly. For each group, analytical (CVA), within-subject (CVI), and between-subject (CVG) coefficients of variability were calculated in addition to percent critical change value (CCV) and index of individuality (IoI). Results: For healthy dogs, calculated NTproBNP values were: CVA ¼ 4.2%; CVI ¼ 25.2%; CVG ¼ 49.3%; IoI ¼ 0.52, and CCV ¼ 70.8%. For dogs with MMVD, calculated NTproBNP values were: CVA ¼ 6.2%; CVI ¼ 20.0%; CVG ¼ 61.3%; IoI ¼ 0.34, and CCV ¼ 58.2%.
The study was performed at the Veterinary Medical Teaching Hospital, Texas A&M University, College Station, TX. The data from this study were presented in oral abstract form at the 2014 ACVIM Forum in Nashville, TN. * Corresponding author. E-mail address: [email protected] (R.L. Winter). http://dx.doi.org/10.1016/j.jvc.2016.11.001 1760-2734/ª 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
2
R.L. Winter et al. Conclusions: Biologic variability affects NTproBNP concentrations in healthy dogs and dogs with MMVD. Monitoring serial individual changes in NTproBNP may be clinically relevant in addition to using population-based reference ranges to determine changes in disease status. ª 2016 Elsevier B.V. All rights reserved.
Abbreviations ACVIM
American College of Veterinary Internal Medicine Ao aorta BV biologic variability CCV critical change value for percent difference in biomarker value CV coefficient of variation CVA analytical coefficient of variation CVG between-subject coefficient of variation CVI within-subject coefficient of variation IoI index of individuality LA left atrium LVID left ventricular internal dimension MMVD myxomatous mitral valve disease NTproBNP N-terminal pro-brain natriuretic peptide VHS vertebral heart size
dogs with MMVD, highlighting its potential value in longitudinal evaluation of this disease [6,8,10]. In humans, NTproBNP has been studied in acquired cardiac disease [11e14], and BV is acknowledged as an important component of interpreting longitudinal changes when assessing disease [15e17]. Knowledge of NTproBNP BV has proven useful in the accurate interpretation of changes in both healthy humans and those with cardiac disease. Evaluation of the BV of NTproBNP in dogs is limited [18,19]. The objectives of this study were to measure NTproBNP in healthy dogs and dogs with MMVD classified into American College of Veterinary Internal Medicine (ACVIM) stages B1, B2, or C-stable in order to estimate BV and calculate the Index of Individuality (IoI) of this analyte to determine if population-based reference ranges are appropriate. A secondary objective was to calculate the critical change value (CCV) for percent difference in NTproBNP concentration needed in order to suggest a change in disease in each group of dogs.
Introduction
Materials and methods
Biologic variability (BV) is defined as the change in analyte concentration that occurs independently of the disease status of the patient [1]. Specifically, this refers to changes in analyte concentration when multiple samples are obtained from either a healthy subject or from a patient with stable disease [1,2]. Treatment recommendations and prognosis frequently depend on disease status, therefore establishing the BV is necessary for understanding the clinical importance of changes in the longitudinal measurement of analytes. Myxomatous mitral valve disease (MMVD) is the most common cardiac disease in dogs and is extensively studied in veterinary medicine [3e5]. Studies in dogs with MMVD have investigated the clinical utility of the circulating cardiac biomarker N-terminal pro-brain natriuretic peptide (NTproBNP) [6e9]. Elevated NTproBNP concentrations have been associated with an increased hazard of congestive heart failure and mortality in
Animals Healthy dogs were prospectively recruited from staff and students, and dogs with MMVD were prospectively recruited from the clinical caseload of the Texas A&M University Veterinary Medical Teaching Hospital from August 2012 to August 2013. The study protocol was approved by the University Institutional Animal Use and Care Committee, and informed consent was obtained from all owners. Inclusion criteria included dogs aged >1 year, weighing between 5.0 and 15.0 kg, without any evidence of clinically important systemic disease. All dogs underwent diagnostic testing immediately prior to entry into the study. Diagnostic tests included physical examination, cardiac auscultation, echocardiography, electrocardiography, indirect blood pressure measurement, serum biochemical analysis, and MMVD dogs additionally had thoracic radiography
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs
3
performed. Auscultation and echocardiography were performed by a cardiology resident (R.L.W.) under supervision by a boarded cardiologist. Dogs were considered healthy if they had no evidence of systemic or cardiac disease. Dogs with MMVD had a characteristic left apical, systolic heart murmur on cardiac auscultation and echocardiographic evidence of mitral valve leaflet thickening and mitral regurgitation. Dogs with MMVD were classified according to the ACVIM guidelines [20]. All dogs with MMVD had echocardiographic findings consistent with a diagnosis of MMVD [21]. Briefly, dogs were classified as stage B1 MMVD if they had no echocardiographic evidence of cardiomegaly even if the VHS was >10.7. To be classified as stage B2, dogs were required to fulfill at least two out of three criteria for cardiomegaly. Dogs were classified as stage B2 if they had evidence of cardiomegaly defined as left atrium to aorta (LA/Ao) ratio in short axis 1.6 [22] or radiographic vertebral heart size (VHS) >10.7 [23]. In addition, all dogs with stage B2 MMVD were required to have a left ventricular internal dimension (LVID) in diastole normalized to body weight 1.7 [24]. The selected cutoff value for normalized LVID in diastole was based on a study that found higher values, although technically within the normal reported range (<1.85), predicted increased risk of death in dogs with MMVD [10]. Dogs classified as stage C-stable had evidence of echocardiographic cardiomegaly and a previous diagnosis of cardiogenic pulmonary edema secondary to MMVD that required standard heart failure therapy, including furosemide, to remain free of clinical signs and remained stable throughout the study. Stable was defined as no clinical or radiographic evidence of cardiogenic pulmonary edema, thus clinically requiring no alteration in medication dosages. For all groups, exclusion criteria included evidence of systemic hypertension (defined in this
study as systolic blood pressure 160 mmHg), a clinically important arrhythmia (defined as requiring anti-arrhythmic medication or pacemaker implantation), significant systemic disease, the presence of congenital cardiac disease, or significant pulmonary hypertension (defined in this study as right ventricular to right atrial pressure gradient 60 mmHg based on echocardiographic measurements of systolic tricuspid regurgitant velocities). All dogs had serum creatinine concentrations within the normal reference range. Systemic blood pressure was indirectly measured by an experienced technician using Doppler ultrasonography and averaging five consecutive measurements on dogs in right lateral recumbency. Dogs with stage C-stable MMVD were excluded if they had evidence of active pulmonary edema or required changes to cardiac medication dosages during the study period.
Blood sampling and analysis Blood was collected from the jugular vein in all dogs. On the first day of evaluation, all diagnostic tests were performed including the serum biochemical analysis. Table 1 shows the blood collection schedule for all dogs. For each healthy dog, blood samples were obtained a total of 16 times over a 6-week period. During week 1, blood samples were collected 2 h 30 min apart within a day, and consistent times were used throughout the week. For weeks 2e6, blood samples were collected on consistent days and at consistent times of the day 45 min. For each MMVD dog, a total of four blood samples were collected at two time points, 7 days apart. Blood samples were collected 2 h 30 min apart within a day, and consistent times were used 45 min between week 1 and 2 collection times. Blood was placed into ethylenediamine tetraacetic acid (EDTA)-containing collection tubes and
Table 1 Blood sample collection schedule for healthy dogs (n ¼ 10) and dogs with myxomatous mitral valve disease (n ¼ 28). Variable
Week 1 Day 1
Tests Healthy MMVD
proBNP x (3) x (2)
Week 2
Week 3
Week 4
Week 5
Week 6
proBNP x x (2)
proBNP x d
proBNP x d
proBNP x d
proBNP x d
Days 2e5 Other x x
proBNP x (2) d
‘x’ indicates that a test was performed once on that specific day or week; ‘x (2)’ indicates that tests were performed twice during that day; ‘x (3)’ indicates that tests were performed three times in that day; ‘other’ represents that diagnostic tests other than NTproBNP measurement occurred on that day. Note: sample collection times were consistent across days by 30 min and across weeks by 45 min. MMVD, myxomatous mitral valve disease; NTproBNP, N-terminal pro-brain natriuretic peptide.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
4
R.L. Winter et al.
immediately centrifuged for 15 min at 1,500 g at room temperature to separate and isolate plasma. Plasma was stored in 500e1,000 mL aliquots in polypropylene tubes. All plasma samples were aliquoted within 30 min of blood collection and stored in 80 C until batch analysis. Plasma samples were packed frozen on dry ice and shipped overnight to IDEXX Laboratories,c and NTproBNP was measured in duplicate with a second-generation canine Cardiopet proBNP testd with demonstrated capability to detect concentrations as low as 16 pmol/L and up to 10,000 pmol/L [25]. All dogs had NTproBNP values within this range.
Statistical analyses Concentrations of NTproBNP were assessed for normality using the Anderson-Darling test using commercially available software.e Descriptive statistics were reported as mean and standard deviation for normally distributed data and as the median and range otherwise. Components of biological variation were calculated separately for each study group by nested analysis of variance models using commercially available statistical software.f Variance values and coefficients of variation were determined for betweensubject (CVG), within-subject (CVI), and analytical (CVA) components of biological variability. The models were structured such that CVG was composed of differences between dogs, CVI was composed of within-day and between-day components, and CVA was derived from the duplicate NTproBNP measurements for each sample. IoI was calculated for each group as: (CV2I þ CVA2)1/2/CVG. The CCV was defined as the change in NTproBNP value needed to suggest disease severity change, and this was estimated based on the method reported by Fraser and Harris [1]: CCV ¼ Z 21/ 2 (CVA2 þ CV2I )1/2, where Z ¼ 1.96 and Z ¼ 2.58 for CCV-95% and CCV-99%, respectively. For each CCV, a 95% confidence interval was calculated by computing nested analysis of variance models for 100 simulated data sets for each group. Simulated data sets were generated using a bootstrapping module of the commercially available statistical softwaref which randomly sampled duplicate
c
IDEXX Laboratories, Westbrook, ME, USA. Cardiopet proBNP second-generation Test-Canine, IDEXX Laboratories, Westbrook, ME, USA. e MINITAB Statistical Software, Release 13.32, Minitab Inc, State College, Pennsylvania, USA. f JMP, Version 11.0.0 SAS Institute Inc., Cary, NC, USA. d
results from a single bleed per dog per day in each simulation.
Results A total of 38 dogs were included in this study, 10 healthy dogs and 28 dogs with MMVD (10 stage B1, 10 stage B2 and eight stage C-stable). Healthy dogs included seven spayed females and three castrated males and had a median weight of 10.7 kg (range, 5.1e13.6 kg), with a median age of 2.3 years (range, 1.3e11.3 years). Healthy dog breeds included two mixed breed and one each of Yorkshire terrier, Papillion, Bichon Frise, Miniature Schnauzer, Border Collie, Beagle, Miniature Australian Shepherd, and Pembroke Welsh Corgi. Dogs with MMVD included 10 spayed females, one intact female, 15 castrated males, and two intact males that had a median weight of 8.3 kg (range, 5.0e11.8 kg) and a median age of 8.9 years (range, 4.1e14.9 years). Age and weight by stage of MMVD are reported in Table 2. Dogs with MMVD included 14 Cavalier King Charles Spaniels, three Miniature Schnauzers, two Chihuahuas, two Miniature Dachshunds, and one each of Cairn Terrier, Yorkshire Terrier, Cocker Spaniel, Jack Russell Terrier, Bichon Frise, American Eskimo dog, and Shetland Sheepdog. Dogs with stage B1 and stage B2 MMVD were not receiving any cardiac medications. All dogs with stage C-stable MMVD were receiving pimobendan and furosemide, and all but one received an angiotensin-converting enzyme inhibitor. All but two dogs with stage C-stable MMVD received spironolactone. One dog with stage C-stable MMVD was receiving carvedilol. Nine dogs (one healthy; five stage B2; three stage C-stable) were receiving additional medications and these included omeprazole (n ¼ 2), denamarin (n ¼ 1), soloxine (n ¼ 1), gabapentin (n ¼ 1), cyclosporine (n ¼ 1), carprofen (n ¼ 1), tramadol (n ¼ 1), diphenhydramine (n ¼ 1), hydrocodone (n ¼ 1), metronidazole (n ¼ 1), famotidine (n ¼ 1), and theophylline (n ¼ 1). No dog was receiving more than two noncardiac medications. Descriptive statistics for echocardiographic and VHS values in healthy dogs and dogs with MMVD are found in Table 2. To further clarify values listed in the table, additional details regarding staging for MMVD dogs are as follows. Dogs with MMVD were categorized as stage B1 based on the absence of echocardiographic cardiomegaly and as stage B2 based on the presence of cardiomegaly. All but one
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs
5
Table 2 Median and range of age, weight, echocardiographic measures and NTproBNP for healthy dogs and dogs with myxomatous mitral valve disease (272 observations on 38 dogs). Variable Age (years) Weight (kg) Echo LA/Ao LVIDdN FS% E/E0 E/IVRT VHS
Healthy
MMVD stage B1
MMVD stage B2
MMVD stage C-stable
N ¼ 10
N ¼ 10
N ¼ 10
N¼8
2.27 (1.32e11.30) 10.65 (5.10e13.6)
8.18 (4.07e11.38) 7.40 (5.60e9.10)
8.16 (6.02e12.32) 8.95 (6.70e11.80)
11.77 (8.99e14.87) 9.15 (5.00e10.70)
1.25 1.39 41.0 8.54 1.76
1.32 (1.13e1.49) 1.46 (1.14e1.64) 37.2 (29.4e49.0) 8.63 (7.13e13.55) 1.11 (0.77e1.61) 10.70 (9.20e11.10)
1.79 (1.35e1.89) 1.82 (1.70e2.31) 46.4 (37.1e51.3) 9.71 (6.05e17.16) 1.61 (1.21e2.67) 11.10 (10.80e12.80)
1.73 (1.61e2.40) 1.86 (1.20e2.26) 48.1 (39.2e64.1) 12.42 (9.13e24.86) 1.81 (1.02e3.90) 12.40 (9.80e12.70)
(1.13e1.42) (1.15e1.59) (27.3e50.5) (6.31e14.60) (0.86e2.50) N ¼ 160
Biomarker NTproBNP (pmol/L)
N ¼ 40
543 (16e1,558)
N ¼ 40
677 (24e1,344)
1,553 (531e3,010)
N ¼ 32 1,963 (424e4,086)
E/E0 , ratio of left ventricular filling velocity to myocardial motion velocity in early diastole; E/IVRT, ratio of early diastolic left ventricular filling velocity to isovolumic relaxation time; FS%, fractional shortening; NTproBNP, N-terminal pro-brain natriuretic peptide pmol/L; LA/Ao, ratio of left atrial size in short axis to aortic diameter; LVIDdN, left ventricular internal dimension in diastole normalized to body weight; VHS, vertebral heart size.
stage B2 dog was classified as stage B2 based on the presence of cardiomegaly defined by an LA/Ao >1.6 and a VHS >10.7. In one stage B2 dog, although the LA/Ao measured 1.35, cardiomegaly was determined based on VHS >10.7 and the presence of subjective echocardiographic evidence of left atrial enlargement that was not represented in the LA/Ao short-axis measurements. All stage B2 dogs had a LVID in diastole normalized to body weight >1.7. In one stage Cstable dog, the VHS was 9.8 but echocardiography confirmed the presence of cardiomegaly and MMVD. For healthy dogs, the median NTproBNP for all samples was 543 pmol/L with a range of 16e1558 pmol/L (Table 2). In healthy dogs, CVA, CVI and CVG were 4.2%, 25.2%, and 49.3%, respectively (Table 3). For all dogs with MMVD, the median NTproBNP for all samples was 991 pmol/L with a range of 24e4086 pmol/L. In all dogs with MMVD (stages B1, B2, and C-stable combined),
CVA, CVI, and CVG were 6.2%, 20.0%, and 61.3%, respectively (Table 3). In dogs with stage B1, B2 and C-stable MMVD the median (ranges) in pmol/L were 677 (24e1,344), 1,553 (531e3,010), and 1,963 (424e4,086), respectively. The CCV-95% (95% confidence interval of critical difference needed to suggest disease severity progression) was 70.8% in healthy dogs and 58.2% in all dogs with MMVD (Table 3). The CCV-99% (99% confidence interval of critical difference needed to suggest disease severity progression) was 93.0% in healthy dogs and 76.5% in all dogs with MMVD (Table 3). The IoI was 0.52 for healthy dogs and 0.34 for all dogs with MMVD (Table 3).
Discussion This study estimated BV of serially measured NTproBNP in healthy dogs and dogs with MMVD. Dogs with MMVD had a lower inherent variability of
Table 3 Analytical variation (CVA), within-subject coefficient of variation (CVI), between-subject coefficient of variation (CVG), Index of Individuality (IoI), and critical change values (both 95% and 99% confidence intervals) for difference in NTproBNP value (CCV) in healthy dogs and dogs with MMVD. Disease state Healthy MMVD B1 MMVD B2 MMVD C-stable All MMVD
Dogs (N)
Samples (N)
CVA (%)
CVI (%)
CVG (%)
IoI
10 10 10 8 28
160 40 40 32 112
4.2 4.4 4.5 7.0 6.2
25.2 26.1 18.0 17.9 20.0
49.3 39.6 35.4 53.8 61.3
0.52 0.69 0.52 0.36 0.34
CCV-95% (95% CI) 70.8% 73.4% 51.4% 53.3% 58.2%
(62.3e82.1%) (64.6e85.2%) (45.2e59.6%) (46.9e61.9%) (51.2e67.5%)
CCV-99% (95% CI) 93.0% 96.5% 67.5% 70.1% 76.5%
(81.8e107.9%) (84.9e111.9%) (59.4e78.3%) (61.7e81.3%) (67.3e88.7%)
CCV, critical change value; CI, confidence interval; MMVD, myxomatous mitral valve disease.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
6 NTproBNP compared to healthy dogs. As a group, dogs with MMVD required a change of 58.2% for NTproBNP to be outside the range of inherent BV and indicate a change in disease. Dogs in this study had NTproBNP measured from plasma samples using the current generation assay platform, which is now commercially available [25]. The increased upper limit of detection for NTproBNP to 10,000 pmol/L (vs. 3,000 in the previous generation), simplified sample submission requirements (plasma or serum) and ongoing research and publications regarding its prognostic utility in dogs with MMVD will likely increase the clinical use of this assay and knowledge of its BV will improve the accuracy of interpretation [6,8,10]. Elevations in NTproBNP can help predict heart failure and survival in dogs with various stages of MMVD [8e10], and dogs with higher NTproBNP values survive a shorter time than those with lower NTproBNP values [6,10]. More recent studies in dogs with MMVD have focused on longitudinal NTproBNP changes to help predict specific clinical end-points such as onset of heart failure or death emphasizing the value of knowing the BV of this analyte in the dog [6,9,10]. Both healthy dogs and dogs with MMVD in this study had IoI values less than 0.6 for NTproBNP. Typically, IoI values <0.6 indicate that conventional population-based reference values are of limited value in the serial evaluation of an individual patient [1,25]. An IoI of >1.4 indicates that population-based reference values are appropriate for interpretation of longitudinal changes in biomarker concentrations in an individual patient [1,26]. Our data suggest that population-based reference ranges for NTproBNP may have limitations, and the incorporation of CCV assessment may be clinically relevant and requires further investigation. Dogs with MMVD stage B2 had a broad range of NTproBNP values that were considered to be within normal BV for this group. The clinical phenotype of dogs with stage B2 MMVD is highly variable in terms of heart size and if and when an individual dog will develop heart failure [3e5,20]. Thus, incorporating knowledge of BV in the interpretation of changes in NTproBNP concentration within one dog sampled multiple times may help better describe these dogs as they progress through the stages of heart disease. It is possible that knowledge of NTproBNP BV in dogs with MMVD stage B2 may help interpret longitudinal samples in this cohort and clinically separate this group of dogs into those which will and those which will not develop heart failure in the near future. Additional longitudinal clinical studies in dogs with MMVD
R.L. Winter et al. stage B2 may help determine if changes greater than the CCV (51.4%) in NTproBNP samples can predict clinically important outcomes such as the likelihood of congestive heart failure or death in a given time frame. As an example, based on the results of this study, an individual dog with MMVD stage B2 and an NTproBNP value of 1,500 pmol/L at one time point would require an NTproBNP value of greater than 2,271 pmol/L at the next time point to indicate a change in disease severity (i.e. 2,271 pmol/L is 51.4% greater than 1,500 pmol/L, the range between these two values comprising the inherent BV of NTproBNP for a dog with MMVD stage B2). The CCV for MMVD as a group as well as individual MMVD stages are provided. Utilizing the CCV for the MMVD group as a whole may benefit a clinician if a dog has a characteristic murmur of MMVD but the heart size is unknown and therefore the exact ACVIM stage is unknown. Because of the variable characteristics between stages of MMVD, if the heart size and stage are known for a dog, then using stage-specific CCV values for BV is relevant. Given the increasing awareness of the clinical utility of cardiac biomarkers such as NTproBNP in veterinary medicine, it is important for veterinarians to have an understanding of how much a biomarker value could change over time due to randomness vs. disease progression in an individual dog [7,27]. Few veterinary studies have assessed the variability of cardiac biomarkers. In one study, NTproBNP was measured in healthy dogs once per week for 3 consecutive weeks, and the majority of dogs had a greater than 100 pmol/L change in NTproBNP values [18]. In a second study, cardiac troponin I (cTnI) and NTproBNP were measured weekly for seven weeks in normal dogs and nine dogs with heart disease not staged with current guidelines [19]. In that study, reduced individual NTproBNP variability was observed in dogs with heart disease compared to control dogs, a result that is consistent with our findings. There are several limitations to this study. The sample size is relatively small; however, there is no standard protocol for determining BV of cardiac biomarkers in human medicine, and BV is often assessed with similarly small numbers of patients [2,15e17]. Additionally, it is possible that individual factors such as age, medications, characteristics of biomarker release or other as yet unknown factors play a role in the BV of biomarkers. Control dogs were recruited from the hospital student and staff population, and therefore, control dogs and dogs with MMVD were not age-matched. Dogs with MMVD were not staged at a second time point, although the time frame of this study was limited
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs to 1 week between evaluations. The time frame for study duration for healthy dogs was chosen based on similar studies in human medicine for determination of BV which ranged from 2 to more than 8 weeks [2,15e17]. Dogs with MMVD were studied for a shorter period of time to reduce the potential for disease progression during the study period. While none of the dogs with MMVD developed clinical signs within the study period, the severity of the heart disease could have changed. Medications were not standardized in the MMVD stage C-stable dogs. A number of factors that may impact NTproBNP BV were not investigated in this study, including breed, medications in stage Cstable dogs, and medications considered unrelated to cardiovascular or systemic health.
Conclusions In conclusion, healthy dogs and dogs with MMVD have inherent BV in the cardiac biomarker NTproBNP. Monitoring serial individual changes in NTproBNP values may be clinically relevant in addition to using population-based reference ranges to determine changes in disease status.
Conflicts of Interest Over the past 5 years, the following authors served as consultants or on advisory boards for and received support (at least one of consulting agreements, speaker honoraria, remuneration of travel expenses, research fundings, programmatic support [intern or resident funding or equipment]) from IDEXX Laboratories: Gordon, Saunders, Winter, Miller. Dr. Jesse S. Buch is an IDEXX Laboratories employee. IDEXX Laboratories did not participate in data collection.
Funding This study was supported by an American Kennel Club Canine Health Foundation Acorn grant (01866-A) and an internal grant at Texas A&M University. IDEXX Laboratories provided sample measurement at no cost and provided statistical support.
Acknowledgments The authors are grateful for the assistance provided by Kathy Glaze, Jill VanWhy, and Dr. May Boggess.
7
References [1] Fraser CG, Harris EK. Generation and application of data on biological variation in clinical chemistry. Crit Rev Clin Lab Sci 1989;27:409e37. [2] Prontera C, Emdin M, Zucchelli GC, Ripoli A, Passino C, Clerico A. Biologic variation of N-terminal pro-brain natriuretic peptide in healthy individuals. Clin Chem 2003;49:1545e55. [3] Borgarelli M, Haggstrom J. Canine degenerative myxomatous mitral valve disease: natural history, clinical presentation and therapy. Vet Clin North Am Small Anim Pract 2010;40:651e63. [4] Borgarelli M, Buchanan JW. Historical review, epidemiology and natural history of degenerative mitral valve disease. J Vet Cardiol 2012;14:93e101. [5] Fox PR. Pathology of myxomatous mitral valve disease in the dog. J Vet Cardiol 2012;14:103e26. [6] Hezzell MJ, Boswood A, Chang YM, Moonarmart W, Souttar K, Elliott J. The combined prognostic potential of serum high-sensitivity cardiac troponin I and N-terminal pro-B-type natriuretic concentrations in dogs with degenerative mitral valve disease. J Vet Intern Med 2012;26: 302e11. [7] Oyama MA, Boswood A, Connolly DJ, Ettinger SJ, Fox PR, Gordon SG, Rush JE, Sisson DD, Stepien RL, Wess G. Clinical usefulness of an assay for measurement of circulating Nterminal pro-B-type natriuretic peptide concentration in dogs and cats with heart disease. J Am Vet Med Assoc 2013; 243:71e82. [8] Wolf J, Gerlack N, Weber K, Klima A, Wess G. Lowered Nterminal pro-B-type natriuretic peptide levels in response to treatment predict survival in dogs with symptomatic mitral valve disease. J Vet Cardiol 2012;14: 399e408. [9] Reynolds CA, Brown DC, Rush JE, Fox PR, Nguyenba TP, Lehmkuhl LB, Gordon SG, Kellihan HB, Stepien RL, Lefbom BK, Meier CK, Oyama MA. Prediction of first onset of congestive heart failure in dogs with degenerative mitral valve disease: the PREDICT cohort study. J Vet Cardiol 2012;14:193e202. [10] Moonamart W, Boswood A, Fuentes VL, Brodbelt D, Souttar K, Elliott J. N-terminal pro B-type natriuretic peptide and left ventricular diameter independently predict mortality in dogs with mitral valve disease. J Small Anim Pract 2010;51:84e96. [11] Gaggin HK, Januzzi Jr JL. Natriuretic peptides in heart failure and acute coronary syndrome. Clin Lab Med 2014; 34:43e58. [12] Verdiani V, Ognibene A, Rutili MS, Lombardo C, Bacci F, Terreni A, Nozzoli C. NT-proBNP reduction percentage during hospital stay predicts long-term mortality and readmission in heart failure patients. J Cardiovasc Med (Hagerstown) 2008;9:694e9. [13] Jourdain P, Jondeau G, Funck F, Gueffet P, Le Helloco A, Donal E, Aupetit JF, Aumont MC, Galinier M, Eicher JC, Cohen-Solal A, Juilliere Y. Plasma brain natriuretic peptide-guided therapy to improve outcome in heart failure: the STARS-BNP multicenter study. J Am Coll Cardiol 2007;49:1733e9. [14] Troughton R, Michael Felker G, Januzzi Jr JL. Natriuretic peptide-guided heart failure management. Eur Heart J 2014;35:16e24. [15] Wu AHB, Smith A, Wieczorek S, Mather JF, Duncan B, White CM, McGill C, Katten D, Heller G. Biological variation for N-terminal pro- and B-type natriuretic peptides and implications for therapeutic monitoring of patients
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
8
[16]
[17]
[18]
[19]
[20]
R.L. Winter et al. with congestive heart failure. Am J Cardiol 2003;92: 628e31. O’Hanlon R, O’Shea P, Ledwidge M, O’Loughlin C, Lange C, Conlon C, Phelan D, Cunningham S, MacDonald K. The biologic variability of B-type natriuretic peptide and Nterminal pro-B-type natriuretic peptide in stable heart failure patients. J Card Fail 2007;13:50e5. Wu AH. Serial testing of B-type natriuretic peptide and NTpro-BNP for monitoring therapy of heart failure: the role of biologic variation in the interpretation of results. Am Heart J 2006;152:828e34. Kellihan HB, Oyama MA, Reynolds CA, Stepien RL. Weekly variability of plasma and serum NT-proBNP measurements in normal dogs. J Vet Cardiol 2009;11:S93e7. Ruaux C, Scollan K, Sucholdolski JS, Steiner JM, Sisson DD. Biologic variability in NT-proBNP and cardiac troponin-I in healthy dogs and dogs with mitral valve degeneration. Vet Clin Pathol 2015;44:420e30. Atkins C, Bonagura J, Ettinger S, Fox P, Gordon S, Haggstrom J, Hamlin R, Keene B, Luis-Fuentes V, Stepien R. Guidelines for the diagnosis and treatment of canine chronic valvular heart disease. J Vet Intern Med 2009;23:1142e50.
[21] Chetboul V, Tissier R. Echocardiographic assessment of canine degenerative mitral valve disease. J Vet Cardiol 2012;14:127e48. [22] Rishniw M, Erb HN. Evaluation of four 2-dimensional echocardiographic methods of assessing left atrial size in dogs. J Vet Intern Med 2000;14:429e35. [23] Buchanan JW, Bucheler J. Vertebral scale system to measure heart size in radiographs. J Am Vet Med Assoc 1995;206:194e9. [24] Cornell CC, Kittleson MD, Della Torre P, Haggstrom J, Lombard CW, Pederson HD, Vollmar A, Wey A. Allometric scaling of M-mode cardiac measurements in normal adult dogs. J Vet Intern Med 2004;18:311e21. [25] Cahill R, Pigeon K, Strong-Townsend M, Drexel JP, Clark GH, Buch JS. Bioanalytical method validation of a second-generation immunoassay for the quantification of NT-proBNP in canine blood. J Vet Intern Med 2013;27:638. [26] Harris EK. Statistical aspects of reference values in clinical pathology. Prog Clin Pathol 1981;8:45e66. [27] Widjaja A, Morris RJ, Levy JC, Frayn KN, Manley SE, Turner RC. Within- and between-subject variation in commonly measured anthropometric and biochemical variables. Clin Chem 1999;45:4561e6.
Available online at www.sciencedirect.com
ScienceDirect
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
-, -e-
www.elsevier.com/locate/jvc
Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease Randolph L. Winter, DVM a,*, Ashley B. Saunders, DVM a, Sonya G. Gordon, DVM, DVSc a, Jesse S. Buch, PhD b, Matthew W. Miller, DVM, MS a a
Department of Small Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX, USA b Immunoassay R&D, IDEXX Laboratories, Westbrook, ME, USA Received 26 June 2016; received in revised form 11 November 2016; accepted 21 November 2016
KEYWORDS Canine; Cardiac biomarker; Mitral regurgitation; Prospective
Abstract Introduction: To determine the biologic variability of N-terminal probrain natriuretic peptide (NTproBNP) in healthy dogs and dogs with various stages of myxomatous mitral valve disease (MMVD). Animals: Thirty-eight privately owned dogs: 28 with MMVD and 10 healthy controls. Materials and methods: Prospective clinical study with comprehensive evaluation used to group dogs as healthy or into three stages of MMVD based on current guidelines. NTproBNP was measured hourly, daily, and weekly. For each group, analytical (CVA), within-subject (CVI), and between-subject (CVG) coefficients of variability were calculated in addition to percent critical change value (CCV) and index of individuality (IoI). Results: For healthy dogs, calculated NTproBNP values were: CVA ¼ 4.2%; CVI ¼ 25.2%; CVG ¼ 49.3%; IoI ¼ 0.52, and CCV ¼ 70.8%. For dogs with MMVD, calculated NTproBNP values were: CVA ¼ 6.2%; CVI ¼ 20.0%; CVG ¼ 61.3%; IoI ¼ 0.34, and CCV ¼ 58.2%.
The study was performed at the Veterinary Medical Teaching Hospital, Texas A&M University, College Station, TX. The data from this study were presented in oral abstract form at the 2014 ACVIM Forum in Nashville, TN. * Corresponding author. E-mail address: [email protected] (R.L. Winter). http://dx.doi.org/10.1016/j.jvc.2016.11.001 1760-2734/ª 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
2
R.L. Winter et al. Conclusions: Biologic variability affects NTproBNP concentrations in healthy dogs and dogs with MMVD. Monitoring serial individual changes in NTproBNP may be clinically relevant in addition to using population-based reference ranges to determine changes in disease status. ª 2016 Elsevier B.V. All rights reserved.
Abbreviations ACVIM
American College of Veterinary Internal Medicine Ao aorta BV biologic variability CCV critical change value for percent difference in biomarker value CV coefficient of variation CVA analytical coefficient of variation CVG between-subject coefficient of variation CVI within-subject coefficient of variation IoI index of individuality LA left atrium LVID left ventricular internal dimension MMVD myxomatous mitral valve disease NTproBNP N-terminal pro-brain natriuretic peptide VHS vertebral heart size
dogs with MMVD, highlighting its potential value in longitudinal evaluation of this disease [6,8,10]. In humans, NTproBNP has been studied in acquired cardiac disease [11e14], and BV is acknowledged as an important component of interpreting longitudinal changes when assessing disease [15e17]. Knowledge of NTproBNP BV has proven useful in the accurate interpretation of changes in both healthy humans and those with cardiac disease. Evaluation of the BV of NTproBNP in dogs is limited [18,19]. The objectives of this study were to measure NTproBNP in healthy dogs and dogs with MMVD classified into American College of Veterinary Internal Medicine (ACVIM) stages B1, B2, or C-stable in order to estimate BV and calculate the Index of Individuality (IoI) of this analyte to determine if population-based reference ranges are appropriate. A secondary objective was to calculate the critical change value (CCV) for percent difference in NTproBNP concentration needed in order to suggest a change in disease in each group of dogs.
Introduction
Materials and methods
Biologic variability (BV) is defined as the change in analyte concentration that occurs independently of the disease status of the patient [1]. Specifically, this refers to changes in analyte concentration when multiple samples are obtained from either a healthy subject or from a patient with stable disease [1,2]. Treatment recommendations and prognosis frequently depend on disease status, therefore establishing the BV is necessary for understanding the clinical importance of changes in the longitudinal measurement of analytes. Myxomatous mitral valve disease (MMVD) is the most common cardiac disease in dogs and is extensively studied in veterinary medicine [3e5]. Studies in dogs with MMVD have investigated the clinical utility of the circulating cardiac biomarker N-terminal pro-brain natriuretic peptide (NTproBNP) [6e9]. Elevated NTproBNP concentrations have been associated with an increased hazard of congestive heart failure and mortality in
Animals Healthy dogs were prospectively recruited from staff and students, and dogs with MMVD were prospectively recruited from the clinical caseload of the Texas A&M University Veterinary Medical Teaching Hospital from August 2012 to August 2013. The study protocol was approved by the University Institutional Animal Use and Care Committee, and informed consent was obtained from all owners. Inclusion criteria included dogs aged >1 year, weighing between 5.0 and 15.0 kg, without any evidence of clinically important systemic disease. All dogs underwent diagnostic testing immediately prior to entry into the study. Diagnostic tests included physical examination, cardiac auscultation, echocardiography, electrocardiography, indirect blood pressure measurement, serum biochemical analysis, and MMVD dogs additionally had thoracic radiography
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs
3
performed. Auscultation and echocardiography were performed by a cardiology resident (R.L.W.) under supervision by a boarded cardiologist. Dogs were considered healthy if they had no evidence of systemic or cardiac disease. Dogs with MMVD had a characteristic left apical, systolic heart murmur on cardiac auscultation and echocardiographic evidence of mitral valve leaflet thickening and mitral regurgitation. Dogs with MMVD were classified according to the ACVIM guidelines [20]. All dogs with MMVD had echocardiographic findings consistent with a diagnosis of MMVD [21]. Briefly, dogs were classified as stage B1 MMVD if they had no echocardiographic evidence of cardiomegaly even if the VHS was >10.7. To be classified as stage B2, dogs were required to fulfill at least two out of three criteria for cardiomegaly. Dogs were classified as stage B2 if they had evidence of cardiomegaly defined as left atrium to aorta (LA/Ao) ratio in short axis 1.6 [22] or radiographic vertebral heart size (VHS) >10.7 [23]. In addition, all dogs with stage B2 MMVD were required to have a left ventricular internal dimension (LVID) in diastole normalized to body weight 1.7 [24]. The selected cutoff value for normalized LVID in diastole was based on a study that found higher values, although technically within the normal reported range (<1.85), predicted increased risk of death in dogs with MMVD [10]. Dogs classified as stage C-stable had evidence of echocardiographic cardiomegaly and a previous diagnosis of cardiogenic pulmonary edema secondary to MMVD that required standard heart failure therapy, including furosemide, to remain free of clinical signs and remained stable throughout the study. Stable was defined as no clinical or radiographic evidence of cardiogenic pulmonary edema, thus clinically requiring no alteration in medication dosages. For all groups, exclusion criteria included evidence of systemic hypertension (defined in this
study as systolic blood pressure 160 mmHg), a clinically important arrhythmia (defined as requiring anti-arrhythmic medication or pacemaker implantation), significant systemic disease, the presence of congenital cardiac disease, or significant pulmonary hypertension (defined in this study as right ventricular to right atrial pressure gradient 60 mmHg based on echocardiographic measurements of systolic tricuspid regurgitant velocities). All dogs had serum creatinine concentrations within the normal reference range. Systemic blood pressure was indirectly measured by an experienced technician using Doppler ultrasonography and averaging five consecutive measurements on dogs in right lateral recumbency. Dogs with stage C-stable MMVD were excluded if they had evidence of active pulmonary edema or required changes to cardiac medication dosages during the study period.
Blood sampling and analysis Blood was collected from the jugular vein in all dogs. On the first day of evaluation, all diagnostic tests were performed including the serum biochemical analysis. Table 1 shows the blood collection schedule for all dogs. For each healthy dog, blood samples were obtained a total of 16 times over a 6-week period. During week 1, blood samples were collected 2 h 30 min apart within a day, and consistent times were used throughout the week. For weeks 2e6, blood samples were collected on consistent days and at consistent times of the day 45 min. For each MMVD dog, a total of four blood samples were collected at two time points, 7 days apart. Blood samples were collected 2 h 30 min apart within a day, and consistent times were used 45 min between week 1 and 2 collection times. Blood was placed into ethylenediamine tetraacetic acid (EDTA)-containing collection tubes and
Table 1 Blood sample collection schedule for healthy dogs (n ¼ 10) and dogs with myxomatous mitral valve disease (n ¼ 28). Variable
Week 1 Day 1
Tests Healthy MMVD
proBNP x (3) x (2)
Week 2
Week 3
Week 4
Week 5
Week 6
proBNP x x (2)
proBNP x d
proBNP x d
proBNP x d
proBNP x d
Days 2e5 Other x x
proBNP x (2) d
‘x’ indicates that a test was performed once on that specific day or week; ‘x (2)’ indicates that tests were performed twice during that day; ‘x (3)’ indicates that tests were performed three times in that day; ‘other’ represents that diagnostic tests other than NTproBNP measurement occurred on that day. Note: sample collection times were consistent across days by 30 min and across weeks by 45 min. MMVD, myxomatous mitral valve disease; NTproBNP, N-terminal pro-brain natriuretic peptide.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
4
R.L. Winter et al.
immediately centrifuged for 15 min at 1,500 g at room temperature to separate and isolate plasma. Plasma was stored in 500e1,000 mL aliquots in polypropylene tubes. All plasma samples were aliquoted within 30 min of blood collection and stored in 80 C until batch analysis. Plasma samples were packed frozen on dry ice and shipped overnight to IDEXX Laboratories,c and NTproBNP was measured in duplicate with a second-generation canine Cardiopet proBNP testd with demonstrated capability to detect concentrations as low as 16 pmol/L and up to 10,000 pmol/L [25]. All dogs had NTproBNP values within this range.
Statistical analyses Concentrations of NTproBNP were assessed for normality using the Anderson-Darling test using commercially available software.e Descriptive statistics were reported as mean and standard deviation for normally distributed data and as the median and range otherwise. Components of biological variation were calculated separately for each study group by nested analysis of variance models using commercially available statistical software.f Variance values and coefficients of variation were determined for betweensubject (CVG), within-subject (CVI), and analytical (CVA) components of biological variability. The models were structured such that CVG was composed of differences between dogs, CVI was composed of within-day and between-day components, and CVA was derived from the duplicate NTproBNP measurements for each sample. IoI was calculated for each group as: (CV2I þ CVA2)1/2/CVG. The CCV was defined as the change in NTproBNP value needed to suggest disease severity change, and this was estimated based on the method reported by Fraser and Harris [1]: CCV ¼ Z 21/ 2 (CVA2 þ CV2I )1/2, where Z ¼ 1.96 and Z ¼ 2.58 for CCV-95% and CCV-99%, respectively. For each CCV, a 95% confidence interval was calculated by computing nested analysis of variance models for 100 simulated data sets for each group. Simulated data sets were generated using a bootstrapping module of the commercially available statistical softwaref which randomly sampled duplicate
c
IDEXX Laboratories, Westbrook, ME, USA. Cardiopet proBNP second-generation Test-Canine, IDEXX Laboratories, Westbrook, ME, USA. e MINITAB Statistical Software, Release 13.32, Minitab Inc, State College, Pennsylvania, USA. f JMP, Version 11.0.0 SAS Institute Inc., Cary, NC, USA. d
results from a single bleed per dog per day in each simulation.
Results A total of 38 dogs were included in this study, 10 healthy dogs and 28 dogs with MMVD (10 stage B1, 10 stage B2 and eight stage C-stable). Healthy dogs included seven spayed females and three castrated males and had a median weight of 10.7 kg (range, 5.1e13.6 kg), with a median age of 2.3 years (range, 1.3e11.3 years). Healthy dog breeds included two mixed breed and one each of Yorkshire terrier, Papillion, Bichon Frise, Miniature Schnauzer, Border Collie, Beagle, Miniature Australian Shepherd, and Pembroke Welsh Corgi. Dogs with MMVD included 10 spayed females, one intact female, 15 castrated males, and two intact males that had a median weight of 8.3 kg (range, 5.0e11.8 kg) and a median age of 8.9 years (range, 4.1e14.9 years). Age and weight by stage of MMVD are reported in Table 2. Dogs with MMVD included 14 Cavalier King Charles Spaniels, three Miniature Schnauzers, two Chihuahuas, two Miniature Dachshunds, and one each of Cairn Terrier, Yorkshire Terrier, Cocker Spaniel, Jack Russell Terrier, Bichon Frise, American Eskimo dog, and Shetland Sheepdog. Dogs with stage B1 and stage B2 MMVD were not receiving any cardiac medications. All dogs with stage C-stable MMVD were receiving pimobendan and furosemide, and all but one received an angiotensin-converting enzyme inhibitor. All but two dogs with stage C-stable MMVD received spironolactone. One dog with stage C-stable MMVD was receiving carvedilol. Nine dogs (one healthy; five stage B2; three stage C-stable) were receiving additional medications and these included omeprazole (n ¼ 2), denamarin (n ¼ 1), soloxine (n ¼ 1), gabapentin (n ¼ 1), cyclosporine (n ¼ 1), carprofen (n ¼ 1), tramadol (n ¼ 1), diphenhydramine (n ¼ 1), hydrocodone (n ¼ 1), metronidazole (n ¼ 1), famotidine (n ¼ 1), and theophylline (n ¼ 1). No dog was receiving more than two noncardiac medications. Descriptive statistics for echocardiographic and VHS values in healthy dogs and dogs with MMVD are found in Table 2. To further clarify values listed in the table, additional details regarding staging for MMVD dogs are as follows. Dogs with MMVD were categorized as stage B1 based on the absence of echocardiographic cardiomegaly and as stage B2 based on the presence of cardiomegaly. All but one
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs
5
Table 2 Median and range of age, weight, echocardiographic measures and NTproBNP for healthy dogs and dogs with myxomatous mitral valve disease (272 observations on 38 dogs). Variable Age (years) Weight (kg) Echo LA/Ao LVIDdN FS% E/E0 E/IVRT VHS
Healthy
MMVD stage B1
MMVD stage B2
MMVD stage C-stable
N ¼ 10
N ¼ 10
N ¼ 10
N¼8
2.27 (1.32e11.30) 10.65 (5.10e13.6)
8.18 (4.07e11.38) 7.40 (5.60e9.10)
8.16 (6.02e12.32) 8.95 (6.70e11.80)
11.77 (8.99e14.87) 9.15 (5.00e10.70)
1.25 1.39 41.0 8.54 1.76
1.32 (1.13e1.49) 1.46 (1.14e1.64) 37.2 (29.4e49.0) 8.63 (7.13e13.55) 1.11 (0.77e1.61) 10.70 (9.20e11.10)
1.79 (1.35e1.89) 1.82 (1.70e2.31) 46.4 (37.1e51.3) 9.71 (6.05e17.16) 1.61 (1.21e2.67) 11.10 (10.80e12.80)
1.73 (1.61e2.40) 1.86 (1.20e2.26) 48.1 (39.2e64.1) 12.42 (9.13e24.86) 1.81 (1.02e3.90) 12.40 (9.80e12.70)
(1.13e1.42) (1.15e1.59) (27.3e50.5) (6.31e14.60) (0.86e2.50) N ¼ 160
Biomarker NTproBNP (pmol/L)
N ¼ 40
543 (16e1,558)
N ¼ 40
677 (24e1,344)
1,553 (531e3,010)
N ¼ 32 1,963 (424e4,086)
E/E0 , ratio of left ventricular filling velocity to myocardial motion velocity in early diastole; E/IVRT, ratio of early diastolic left ventricular filling velocity to isovolumic relaxation time; FS%, fractional shortening; NTproBNP, N-terminal pro-brain natriuretic peptide pmol/L; LA/Ao, ratio of left atrial size in short axis to aortic diameter; LVIDdN, left ventricular internal dimension in diastole normalized to body weight; VHS, vertebral heart size.
stage B2 dog was classified as stage B2 based on the presence of cardiomegaly defined by an LA/Ao >1.6 and a VHS >10.7. In one stage B2 dog, although the LA/Ao measured 1.35, cardiomegaly was determined based on VHS >10.7 and the presence of subjective echocardiographic evidence of left atrial enlargement that was not represented in the LA/Ao short-axis measurements. All stage B2 dogs had a LVID in diastole normalized to body weight >1.7. In one stage Cstable dog, the VHS was 9.8 but echocardiography confirmed the presence of cardiomegaly and MMVD. For healthy dogs, the median NTproBNP for all samples was 543 pmol/L with a range of 16e1558 pmol/L (Table 2). In healthy dogs, CVA, CVI and CVG were 4.2%, 25.2%, and 49.3%, respectively (Table 3). For all dogs with MMVD, the median NTproBNP for all samples was 991 pmol/L with a range of 24e4086 pmol/L. In all dogs with MMVD (stages B1, B2, and C-stable combined),
CVA, CVI, and CVG were 6.2%, 20.0%, and 61.3%, respectively (Table 3). In dogs with stage B1, B2 and C-stable MMVD the median (ranges) in pmol/L were 677 (24e1,344), 1,553 (531e3,010), and 1,963 (424e4,086), respectively. The CCV-95% (95% confidence interval of critical difference needed to suggest disease severity progression) was 70.8% in healthy dogs and 58.2% in all dogs with MMVD (Table 3). The CCV-99% (99% confidence interval of critical difference needed to suggest disease severity progression) was 93.0% in healthy dogs and 76.5% in all dogs with MMVD (Table 3). The IoI was 0.52 for healthy dogs and 0.34 for all dogs with MMVD (Table 3).
Discussion This study estimated BV of serially measured NTproBNP in healthy dogs and dogs with MMVD. Dogs with MMVD had a lower inherent variability of
Table 3 Analytical variation (CVA), within-subject coefficient of variation (CVI), between-subject coefficient of variation (CVG), Index of Individuality (IoI), and critical change values (both 95% and 99% confidence intervals) for difference in NTproBNP value (CCV) in healthy dogs and dogs with MMVD. Disease state Healthy MMVD B1 MMVD B2 MMVD C-stable All MMVD
Dogs (N)
Samples (N)
CVA (%)
CVI (%)
CVG (%)
IoI
10 10 10 8 28
160 40 40 32 112
4.2 4.4 4.5 7.0 6.2
25.2 26.1 18.0 17.9 20.0
49.3 39.6 35.4 53.8 61.3
0.52 0.69 0.52 0.36 0.34
CCV-95% (95% CI) 70.8% 73.4% 51.4% 53.3% 58.2%
(62.3e82.1%) (64.6e85.2%) (45.2e59.6%) (46.9e61.9%) (51.2e67.5%)
CCV-99% (95% CI) 93.0% 96.5% 67.5% 70.1% 76.5%
(81.8e107.9%) (84.9e111.9%) (59.4e78.3%) (61.7e81.3%) (67.3e88.7%)
CCV, critical change value; CI, confidence interval; MMVD, myxomatous mitral valve disease.
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
6 NTproBNP compared to healthy dogs. As a group, dogs with MMVD required a change of 58.2% for NTproBNP to be outside the range of inherent BV and indicate a change in disease. Dogs in this study had NTproBNP measured from plasma samples using the current generation assay platform, which is now commercially available [25]. The increased upper limit of detection for NTproBNP to 10,000 pmol/L (vs. 3,000 in the previous generation), simplified sample submission requirements (plasma or serum) and ongoing research and publications regarding its prognostic utility in dogs with MMVD will likely increase the clinical use of this assay and knowledge of its BV will improve the accuracy of interpretation [6,8,10]. Elevations in NTproBNP can help predict heart failure and survival in dogs with various stages of MMVD [8e10], and dogs with higher NTproBNP values survive a shorter time than those with lower NTproBNP values [6,10]. More recent studies in dogs with MMVD have focused on longitudinal NTproBNP changes to help predict specific clinical end-points such as onset of heart failure or death emphasizing the value of knowing the BV of this analyte in the dog [6,9,10]. Both healthy dogs and dogs with MMVD in this study had IoI values less than 0.6 for NTproBNP. Typically, IoI values <0.6 indicate that conventional population-based reference values are of limited value in the serial evaluation of an individual patient [1,25]. An IoI of >1.4 indicates that population-based reference values are appropriate for interpretation of longitudinal changes in biomarker concentrations in an individual patient [1,26]. Our data suggest that population-based reference ranges for NTproBNP may have limitations, and the incorporation of CCV assessment may be clinically relevant and requires further investigation. Dogs with MMVD stage B2 had a broad range of NTproBNP values that were considered to be within normal BV for this group. The clinical phenotype of dogs with stage B2 MMVD is highly variable in terms of heart size and if and when an individual dog will develop heart failure [3e5,20]. Thus, incorporating knowledge of BV in the interpretation of changes in NTproBNP concentration within one dog sampled multiple times may help better describe these dogs as they progress through the stages of heart disease. It is possible that knowledge of NTproBNP BV in dogs with MMVD stage B2 may help interpret longitudinal samples in this cohort and clinically separate this group of dogs into those which will and those which will not develop heart failure in the near future. Additional longitudinal clinical studies in dogs with MMVD
R.L. Winter et al. stage B2 may help determine if changes greater than the CCV (51.4%) in NTproBNP samples can predict clinically important outcomes such as the likelihood of congestive heart failure or death in a given time frame. As an example, based on the results of this study, an individual dog with MMVD stage B2 and an NTproBNP value of 1,500 pmol/L at one time point would require an NTproBNP value of greater than 2,271 pmol/L at the next time point to indicate a change in disease severity (i.e. 2,271 pmol/L is 51.4% greater than 1,500 pmol/L, the range between these two values comprising the inherent BV of NTproBNP for a dog with MMVD stage B2). The CCV for MMVD as a group as well as individual MMVD stages are provided. Utilizing the CCV for the MMVD group as a whole may benefit a clinician if a dog has a characteristic murmur of MMVD but the heart size is unknown and therefore the exact ACVIM stage is unknown. Because of the variable characteristics between stages of MMVD, if the heart size and stage are known for a dog, then using stage-specific CCV values for BV is relevant. Given the increasing awareness of the clinical utility of cardiac biomarkers such as NTproBNP in veterinary medicine, it is important for veterinarians to have an understanding of how much a biomarker value could change over time due to randomness vs. disease progression in an individual dog [7,27]. Few veterinary studies have assessed the variability of cardiac biomarkers. In one study, NTproBNP was measured in healthy dogs once per week for 3 consecutive weeks, and the majority of dogs had a greater than 100 pmol/L change in NTproBNP values [18]. In a second study, cardiac troponin I (cTnI) and NTproBNP were measured weekly for seven weeks in normal dogs and nine dogs with heart disease not staged with current guidelines [19]. In that study, reduced individual NTproBNP variability was observed in dogs with heart disease compared to control dogs, a result that is consistent with our findings. There are several limitations to this study. The sample size is relatively small; however, there is no standard protocol for determining BV of cardiac biomarkers in human medicine, and BV is often assessed with similarly small numbers of patients [2,15e17]. Additionally, it is possible that individual factors such as age, medications, characteristics of biomarker release or other as yet unknown factors play a role in the BV of biomarkers. Control dogs were recruited from the hospital student and staff population, and therefore, control dogs and dogs with MMVD were not age-matched. Dogs with MMVD were not staged at a second time point, although the time frame of this study was limited
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
Biologic variability of NTproBNP in dogs to 1 week between evaluations. The time frame for study duration for healthy dogs was chosen based on similar studies in human medicine for determination of BV which ranged from 2 to more than 8 weeks [2,15e17]. Dogs with MMVD were studied for a shorter period of time to reduce the potential for disease progression during the study period. While none of the dogs with MMVD developed clinical signs within the study period, the severity of the heart disease could have changed. Medications were not standardized in the MMVD stage C-stable dogs. A number of factors that may impact NTproBNP BV were not investigated in this study, including breed, medications in stage Cstable dogs, and medications considered unrelated to cardiovascular or systemic health.
Conclusions In conclusion, healthy dogs and dogs with MMVD have inherent BV in the cardiac biomarker NTproBNP. Monitoring serial individual changes in NTproBNP values may be clinically relevant in addition to using population-based reference ranges to determine changes in disease status.
Conflicts of Interest Over the past 5 years, the following authors served as consultants or on advisory boards for and received support (at least one of consulting agreements, speaker honoraria, remuneration of travel expenses, research fundings, programmatic support [intern or resident funding or equipment]) from IDEXX Laboratories: Gordon, Saunders, Winter, Miller. Dr. Jesse S. Buch is an IDEXX Laboratories employee. IDEXX Laboratories did not participate in data collection.
Funding This study was supported by an American Kennel Club Canine Health Foundation Acorn grant (01866-A) and an internal grant at Texas A&M University. IDEXX Laboratories provided sample measurement at no cost and provided statistical support.
Acknowledgments The authors are grateful for the assistance provided by Kathy Glaze, Jill VanWhy, and Dr. May Boggess.
7
References [1] Fraser CG, Harris EK. Generation and application of data on biological variation in clinical chemistry. Crit Rev Clin Lab Sci 1989;27:409e37. [2] Prontera C, Emdin M, Zucchelli GC, Ripoli A, Passino C, Clerico A. Biologic variation of N-terminal pro-brain natriuretic peptide in healthy individuals. Clin Chem 2003;49:1545e55. [3] Borgarelli M, Haggstrom J. Canine degenerative myxomatous mitral valve disease: natural history, clinical presentation and therapy. Vet Clin North Am Small Anim Pract 2010;40:651e63. [4] Borgarelli M, Buchanan JW. Historical review, epidemiology and natural history of degenerative mitral valve disease. J Vet Cardiol 2012;14:93e101. [5] Fox PR. Pathology of myxomatous mitral valve disease in the dog. J Vet Cardiol 2012;14:103e26. [6] Hezzell MJ, Boswood A, Chang YM, Moonarmart W, Souttar K, Elliott J. The combined prognostic potential of serum high-sensitivity cardiac troponin I and N-terminal pro-B-type natriuretic concentrations in dogs with degenerative mitral valve disease. J Vet Intern Med 2012;26: 302e11. [7] Oyama MA, Boswood A, Connolly DJ, Ettinger SJ, Fox PR, Gordon SG, Rush JE, Sisson DD, Stepien RL, Wess G. Clinical usefulness of an assay for measurement of circulating Nterminal pro-B-type natriuretic peptide concentration in dogs and cats with heart disease. J Am Vet Med Assoc 2013; 243:71e82. [8] Wolf J, Gerlack N, Weber K, Klima A, Wess G. Lowered Nterminal pro-B-type natriuretic peptide levels in response to treatment predict survival in dogs with symptomatic mitral valve disease. J Vet Cardiol 2012;14: 399e408. [9] Reynolds CA, Brown DC, Rush JE, Fox PR, Nguyenba TP, Lehmkuhl LB, Gordon SG, Kellihan HB, Stepien RL, Lefbom BK, Meier CK, Oyama MA. Prediction of first onset of congestive heart failure in dogs with degenerative mitral valve disease: the PREDICT cohort study. J Vet Cardiol 2012;14:193e202. [10] Moonamart W, Boswood A, Fuentes VL, Brodbelt D, Souttar K, Elliott J. N-terminal pro B-type natriuretic peptide and left ventricular diameter independently predict mortality in dogs with mitral valve disease. J Small Anim Pract 2010;51:84e96. [11] Gaggin HK, Januzzi Jr JL. Natriuretic peptides in heart failure and acute coronary syndrome. Clin Lab Med 2014; 34:43e58. [12] Verdiani V, Ognibene A, Rutili MS, Lombardo C, Bacci F, Terreni A, Nozzoli C. NT-proBNP reduction percentage during hospital stay predicts long-term mortality and readmission in heart failure patients. J Cardiovasc Med (Hagerstown) 2008;9:694e9. [13] Jourdain P, Jondeau G, Funck F, Gueffet P, Le Helloco A, Donal E, Aupetit JF, Aumont MC, Galinier M, Eicher JC, Cohen-Solal A, Juilliere Y. Plasma brain natriuretic peptide-guided therapy to improve outcome in heart failure: the STARS-BNP multicenter study. J Am Coll Cardiol 2007;49:1733e9. [14] Troughton R, Michael Felker G, Januzzi Jr JL. Natriuretic peptide-guided heart failure management. Eur Heart J 2014;35:16e24. [15] Wu AHB, Smith A, Wieczorek S, Mather JF, Duncan B, White CM, McGill C, Katten D, Heller G. Biological variation for N-terminal pro- and B-type natriuretic peptides and implications for therapeutic monitoring of patients
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001
8
[16]
[17]
[18]
[19]
[20]
R.L. Winter et al. with congestive heart failure. Am J Cardiol 2003;92: 628e31. O’Hanlon R, O’Shea P, Ledwidge M, O’Loughlin C, Lange C, Conlon C, Phelan D, Cunningham S, MacDonald K. The biologic variability of B-type natriuretic peptide and Nterminal pro-B-type natriuretic peptide in stable heart failure patients. J Card Fail 2007;13:50e5. Wu AH. Serial testing of B-type natriuretic peptide and NTpro-BNP for monitoring therapy of heart failure: the role of biologic variation in the interpretation of results. Am Heart J 2006;152:828e34. Kellihan HB, Oyama MA, Reynolds CA, Stepien RL. Weekly variability of plasma and serum NT-proBNP measurements in normal dogs. J Vet Cardiol 2009;11:S93e7. Ruaux C, Scollan K, Sucholdolski JS, Steiner JM, Sisson DD. Biologic variability in NT-proBNP and cardiac troponin-I in healthy dogs and dogs with mitral valve degeneration. Vet Clin Pathol 2015;44:420e30. Atkins C, Bonagura J, Ettinger S, Fox P, Gordon S, Haggstrom J, Hamlin R, Keene B, Luis-Fuentes V, Stepien R. Guidelines for the diagnosis and treatment of canine chronic valvular heart disease. J Vet Intern Med 2009;23:1142e50.
[21] Chetboul V, Tissier R. Echocardiographic assessment of canine degenerative mitral valve disease. J Vet Cardiol 2012;14:127e48. [22] Rishniw M, Erb HN. Evaluation of four 2-dimensional echocardiographic methods of assessing left atrial size in dogs. J Vet Intern Med 2000;14:429e35. [23] Buchanan JW, Bucheler J. Vertebral scale system to measure heart size in radiographs. J Am Vet Med Assoc 1995;206:194e9. [24] Cornell CC, Kittleson MD, Della Torre P, Haggstrom J, Lombard CW, Pederson HD, Vollmar A, Wey A. Allometric scaling of M-mode cardiac measurements in normal adult dogs. J Vet Intern Med 2004;18:311e21. [25] Cahill R, Pigeon K, Strong-Townsend M, Drexel JP, Clark GH, Buch JS. Bioanalytical method validation of a second-generation immunoassay for the quantification of NT-proBNP in canine blood. J Vet Intern Med 2013;27:638. [26] Harris EK. Statistical aspects of reference values in clinical pathology. Prog Clin Pathol 1981;8:45e66. [27] Widjaja A, Morris RJ, Levy JC, Frayn KN, Manley SE, Turner RC. Within- and between-subject variation in commonly measured anthropometric and biochemical variables. Clin Chem 1999;45:4561e6.
Available online at www.sciencedirect.com
ScienceDirect
Please cite this article in press as: Winter RL, et al., Biologic variability of N-terminal pro-brain natriuretic peptide in healthy dogs and dogs with myxomatous mitral valve disease, Journal of Veterinary Cardiology (2016), http://dx.doi.org/10.1016/ j.jvc.2016.11.001