Correct interpretation of blood pressure (BP) and heart rate (HR) recordings is important in a clinical environment, but little is known about effects of stress on BP and HR responses of dogs to different clinical settings.
Correct interpretation of blood pressure (BP) and heart rate (HR) recordings is important in a clinical environment, but little is known about effects of stress on BP and HR responses of dogs to different clinical settings.
To investigate BP and HR responses in different clinical settings in dogs of 3 breeds, and to relate findings to urinary catecholamine concentrations measured by ELISA assays previously validated for use in human plasma and urine, after validation for use in dogs.
Client-owned healthy dogs; 41 Labrador Retrievers, 33 Cavalier King Charles Spaniels (CKCS), and 15 Dachshunds.
Prospective observational study. BP and HR were measured in 4 clinical settings with or without veterinarian and owner present. Urine samples were taken before and after examination. ELISA assays were validated for canine urine, and epinephrine/creatinine and norepinephrine/creatinine ratios were analyzed.
BP and HR were higher when measured by veterinarian alone than when owner was present (P < .020). Urinary catecholamine/creatinine ratios were higher after examination, compared with before, in all dogs (P < .0001). Labrador Retrievers had lower diastolic BP than Dachshunds in 2 settings (P ≤ .041), lower HR than CKCSs in 3 settings (all P < .0001), and lower catecholamine/creatinine ratios after examination than both other breeds (P ≤ .035). The in-house validation showed mean spiked recovery of 96.5% for epinephrine and 83.8% for norepinephrine.
BP and HR responses were related to breed as well as clinical setting. Breed differences were detected in urinary catecholamine/creatinine ratios. Further studies on breed differences are warranted.
Cavalier King Charles Spaniel
coefficient of variation
diastolic blood pressure
high-pressure liquid chromatography
systolic blood pressure
Recognition and diagnosis of systemic hypertension are important to initiate treatment when indicated, thereby reducing risk of future target-organ damage. Blood pressure (BP) recordings are performed routinely at many veterinary clinics. Several factors can, however, affect results of these recordings. Regulation of BP is complex and involves several organ systems and hormones. It is also influenced by emotional factors, and it is well known that stress can induce high BP in some healthy human patients.[1, 2] In a study on healthy cats, arrival to a waiting room and handling during clinical examination were associated with significant increases in BP. Less is known about stress-induced effects on BP and heart rate (HR) in dogs in different clinical settings. Blood pressure has been shown to differ among breeds. Conflicting results exist concerning breed and body size effects on HR,[4-6] and temperament may play a great role.
Excitement or anxiety activates the sympathetic nervous system and leads to increased catecholamine release from the adrenal medulla and increased output of norepinephrine from adrenergic nerve endings. Circulating catecholamines have a half-life of only few minutes, while human urinary catecholamines are stable, with highest stability found in acidified urine.[8, 9] Analysis of catecholamine concentrations has been performed by high-pressure liquid chromatography (HPLC) in canine urine.[10-13] This is, however, a complicated, expensive, and time-consuming method. Measurement of catecholamine concentrations by ELISA methods in human urine has been found simple, rapid, and accurate. It was therefore of interest to investigate if human ELISA assays for urinary epinephrine and norepinephrine could be adapted and validated for use in canine urine.
The aims of the present study were to investigate BP and HR responses in different clinical settings in dogs of three breeds, and to relate findings to urinary catecholamine concentrations measured by human ELISA assays, after validation for use in dogs.
The study was approved by the Local Ethical Committee in Uppsala, Sweden. To be included, dogs had to be intact healthy males, aged 2–5 years, with a pedigree belonging to the Cavalier King Charles Spaniel (CKCS), Dachshund, or Labrador Retriever breed. Included dogs had to be fed a standard commercial dog food with controlled levels of electrolytes and energy content. Dogs were privately owned and informed owner consent was obtained. Exclusion criteria consisted of any finding indicating systemic or organ-related disease.
To avoid excessive salt intake, owners were instructed not to give any other food or treats than their standard dog food for 2 weeks before participation in the study. A package containing urine sampling material and a piece of elastic band was sent to each owner. They were instructed to practice urine sampling at home at least 3 times and to wrap the elastic band around the base of tail of their dogs and leave it there for 30 minutes a day, during 3 days, to accustom dogs to urine sampling and blood pressure measurements.
On examination day, dogs were fasted and water bowl was removed from 7 am. Voided morning urine samples were collected by owners at home and separated into 1 tube without additives and another tube containing HCl for acidification. Morning urine samples brought in by owners were examined by standard urine analysis and frozen for analysis of catecholamines (acidified sample) and creatinine (nonacidified sample).
The examination of each dog started at 10 am at the clinic (Djurdoktorn Animal Practice, Västerås, Sweden). All examinations were performed by the same experienced small animal veterinarian according to a standardized protocol. Dogs were brought directly from the car to a quiet examination room. They were allowed to adapt to the room for 10–15 minutes together with owners. Thereafter, an appropriate cuff (width 40% of tail circumference)[16, 17] was applied to the base of tail on standing dogs. Blood pressure was indirectly measured with an automated high-definition oscillometric (HDO) device1 in 4 settings: (1) Standard setting: Dog lightly restrained by owner. Once consistent consecutive readings were obtained, veterinarian made 6 recordings. (2) Owner alone: After instructing the owner, veterinarian left the room and owner alone made 4 recordings. (3) Veterinarian alone: Owner left the room and veterinarian alone made 2 recordings. (4) Squeaky toy: With owner still outside the room, veterinarian made another 2 recordings while alerting the dog with a squeaky toy. At least 10 seconds was allowed between all cuff inflations. To measure the immediate response to settings 3 and 4, only 2 recordings were made. All dogs were examined in the same order starting with the standard setting and ending with the setting assumed most exciting, ie, the squeaky toy. In all settings, systolic (SBP) and diastolic (DBP) blood pressure and HR were recorded from the HDO device. No exclusion of readings was made during recordings. The HDO device, which indirectly measures peripheral blood pressure, has been compared with directly measured central arterial blood pressure with telemetry.[18, 19] In the present study, the HDO device was tested for repeatability by 6 recordings in each dog during the standard setting. The coefficient of variation (CV) ranged between 2 and 16% (SBP), 2 and 39% (DBP), and 1 and 38% (HR).
All dogs underwent general physical examination, 5-minute ECG registration,2 and echocardiographic examination. The echocardiographic examination was performed with dog in right and left lateral recumbency with an ultrasonographic unit3 equipped with an 8–3 MHz phased-array transducer for small dogs and 3–1 MHz phased-array transducer for large-breed dogs. Continuous ECG monitoring was performed and images and loops from standardized imaging planes were digitally stored. M-mode measurements of the left ventricle were made with standard techniques. The left atrial to aortic root ratio was quantified from the right two-dimensional short-axis view. Pulmonic flow velocity was measured by pulsed-wave Doppler and aortic flow velocity was measured by subcostal continuous wave Doppler. The mitral, tricuspid, aortic, and pulmonic valves were screened with color-flow Doppler.
Blood sampling by venipuncture and analyses of hematology and biochemistry (liver and kidney variables, glucose, and electrolytes) were performed.
After health examination, another voided urine sample was collected by owners outside the clinic, and frozen for analysis of catecholamines (acidified urine) and creatinine (non-acidified urine). Frozen urine was kept at −20°C for a maximum of 3 weeks, transferred frozen to −80°C, and stored for batched analysis. Before analysis, frozen urine was thawed slowly at room temperature and assayed in duplicate for both catecholamine (acidified urine) and creatinine (nonacidified urine) concentrations according to manufacturer's instructions.
Epinephrine and norepinephrine have recently been analyzed in canine urine by HPLC.[10-12] No disturbing substances interfering with the analyses were found in these studies. Therefore human ELISA4 assays for epinephrine and norepinephrine analyses in canine urine were tested in the present study.
The in-house validation was performed by an experienced laboratory technician. The same technician, who was blinded to the experimental design, also analyzed all samples from the dogs. The recovery of epinephrine in spiked urine samples was on average 96.5% (range 91.6–105.2%) and of norepinephrine 83.8% (range 71.6–107.7%). For epinephrine, the lower detection limit was 4.1 nmol/L and the dynamic range was 29–819 nmol/L. The corresponding values for norepinephrine were 19.6 nmol/L and 27–2955 nmol/L.
Interassay CV was calculated from results of the high and low control samples, respectively, which were included in each of the 5 assays, giving a mean CV of 21.5% (at 34.1 nmol/L) and 17.2% (at 169.3 nmol/L) for epinephrine, and of 12.7% (at 145.1 nmol/L) and 15.5% (at 739.4 nmol/L) for norepinephrine. Intraassay CV was calculated as the difference between duplicates using a precision profile, ie, the smoothed relationship of concentration error expressed as %CV value versus concentration giving a CV < 12% (values between 29.0 and 819 nmol/L) for epinephrine and < 14% (values between 26.8 and 2955 nmol/L) for norepinephrine. Canine urinary catecholamines have been proven stable when kept frozen. As the present study compares samples taken before and after the examination in the same dog, they were always analyzed in the same batch and all samples were treated according to the same protocol. In each sample, maximum 200 μL of HCl was added to 3 mL of urine, resulting in dilution of samples by 6,7% and pH values of 2–3.
Concentration of creatinine was analyzed with a commercially available ELISA assay,5 validated for canine urine. Lower detection limit was 0.43 μmol/L, interassay CV was 4.6% (at 121.0 μmol) and 2.4% (at 906.7 μmol/L), and intraassay CV was < 10% (values between 1.44–1768 μmol/L).
Although not required for the main study, owners of the breed with the largest number of participating dogs, ie, Labrador Retrievers, were offered to participate in a home examination, performed within 3 months after examination at the clinic. Preparations were made by owners as described above. Upon arrival of the veterinarian at 11 am, the morning urine sample was examined by standard urine analysis and BP recordings were performed in the same 4 settings as described above. A 2nd urine sample was collected by owners outside their homes, transferred to tubes as described above, and all samples were frozen for later batched analysis.
Commercially available software6 was used for statistical analyses. Mean values for SBP, DBP, and HR were determined for each dog in each setting, after discarding recordings deviating > 20% in SBP. These individual mean values were used to determine mean ± SD values and ranges for SBP, DBP, and HR in each of the 4 settings in all dogs. As the same dogs were examined in different settings, data were analyzed with dog as a random factor in a mixed-procedure model. Comparisons between settings were studied by constructing appropriate contrasts using a variance components covariance structure. With the same method, the clinical values of SBP, DBP, and HR between the 28 Labrador Retrievers examined only at the clinic and the 13 Labrador Retrievers examined also at home were compared to verify that the 13 dogs were representative of the larger population. Potential associations between HR, and SBP and DBP, respectively, in the standard setting were investigated with linear regression analysis.
Urinary catecholamine concentrations were assessed as ratios to creatinine concentrations and are presented as mean ± SD values and ranges. Logarithmic transformation was conducted to correct non-normality for epinephrine/creatinine and norepinephrine/creatinine ratios, and a mixed model with dog as a random factor and breed and sample as fixed factors was used. Analysis of covariance (ANCOVA) with the Tukey post-hoc test was used to compare pre-post differences in ratios between breeds, where breed was modeled as a fixed variable and baseline ratio value as covariate. Potential associations between epinephrine/creatinine and norepinephrine/creatinine ratios after examination, and SBP, DBP and HR, respectively, in the standard setting, were investigated with linear regression analysis. The minimal level of significance was set at P < .05 and Bonferroni adjustment was used.
Ninety-three privately owned dogs were examined. Four dogs were excluded: 3 dogs because of myxomatous mitral valve disease and 1 dog because of extreme stress making completion of protocol impossible. In remaining dogs, no abnormal findings were made during general physical examination, no ECG abnormalities were detected and all echocardiographic variables were within reference values. Furthermore, no clinically significant abnormalities were detected in the hematological and biochemistry variables. Eighty-nine healthy male dogs were therefore included: 41 Labrador Retrievers (mean ± SD age 42.2 ± 13.1 months; body weight (BW) 31.4 ± 4.9 kg), 33 CKCSs (age 38.6 ± 12.7 months; BW 9.5 ± 1.7 kg), and 15 Dachshunds (age 40.7 ± 13.4 months; BW 10.3 ± 1.3 kg).
The SBP, DBP, and HR in all dogs are presented in Table 1. Ranges were wide in all settings for all variables, with individual dogs having BP values up to 180/109 mmHg in standard setting and 200/141 mmHg in veterinarian alone setting. Low values were also observed in some dogs, down to 105 mmHg in SBP and 40 mmHg in DBP in several settings, Table 1. The SBP, DBP, and HR within each breed are presented in Figure 1.
|Standard Setting||Owner Alone||Veterinarian Alone||Squeaky Toy|
|SBP (mmHg)|| |
139 ± 16a
137 ± 16a
144 ± 19b
141 ± 19ab
|DBP (mmHg)|| |
71 ± 13a
72 ± 13ab
76 ± 16b
75 ± 15b
|HR (beats/min)|| |
105 ± 26a
104 ± 29a
106 ± 30a
111 ± 28b
The SBP was higher when measured by veterinarian alone, compared with standard (P = .006) and owner (P = .0002) settings; DBP was higher in veterinarian alone (P = .0044) and squeaky toy (P = .0087) settings compared with the standard setting; and HR was higher in the squeaky toy setting, compared with all other three settings (standard setting P = .0193; owner alone P = .0007; and veterinarian alone P = .030), Table 1.
Labrador Retrievers showed higher SBP when measured by veterinarian alone, compared with standard (P = .020) and owner (P = .025) settings; DBP was higher when measured by veterinarian alone, compared with standard setting (P = .0023); and HR was higher when dogs were alerted with squeaky toy, compared with all other three settings (standard setting P = .0056; owner alone P = .0005; and veterinarian alone P = .0007). Within the other 2 breeds, no differences were found among settings.
Labrador Retrievers had lower DBP compared with Dachshunds in standard (P = .022) and owner (P = .041) settings and lower HR compared with CKCS in the first 3 settings (all P < .0001), Figure 1.
There was a weak, but positive, association between HR and SBP (R2 = 0.07, P = .011), and DBP (R2 = 0.29, P < .0001), respectively, in all settings.
Thirteen of the 41 Labrador Retriever owners agreed to participate in home measurements of their dogs. No differences were found in SBP, DBP, or HR between measurements made at home and at clinic, Table 2. No differences were found between values of SBP, DBP, and HR obtained at examination at the clinic, between the 28 Labrador Retrievers examined only at the clinic and the 13 Labrador Retrievers examined also at home.
|Setting||SBP (mmHg)||DBP (mmHg)||HR (beats/min)|
|Standard setting|| |
134 ± 16
133 ± 21
65 ± 16
61 ± 17
98 ± 25
88 ± 21
|Owner alone|| |
131 ± 16
134 ± 18
68 ± 13
66 ± 14
94 ± 25
84 ± 13
|Veterinarian alone|| |
136 ± 17
140 ± 22
68 ± 17
66 ± 9
99 ± 28
83 ± 17
|Squeaky toy|| |
136 ± 20
134 ± 21
71 ± 13
67 ± 17
98 ± 27
104 ± 44
Epinephrine/creatinine and norepinephrine/creatinine ratios in all dogs and in the different breeds are presented in Table 3.
|Epinephrine/Creatinine Ratio (nmol/mmol)||Difference Before – After||Adj P-Value||Norepinephrine/Creatinine Ratio (nmol/mmol)||Difference Before – After||Adj P-Value|
|All dogs|| |
2.2 ± 1.9
9.0 ± 9.6
6.8 ± 8.7
6.0 ± 3.5
10.5 ± 7.4
4.5 ± 6.0
|Labrador Retriever|| |
1.4 ± 0.7a
3.8 ± 2.4a
2.4 ± 2.2
4.1 ± 2.4a
6.3 ± 3.1a
2.2 ± 2.4
2.8 ± 2.7a
13.7 ± 13.6b
10.9 ± 12.5
7.1 ± 3.5b
13.7 ± 6.0b
6.6 ± 4.8
2.8 ± 1.6a
12.4 ± 6.9b
9.6 ± 6.5
8.1 ± 3.7b
14.3 ± 10.5b
6.2 ± 9.9
Comparing epinephrine/creatinine ratios, higher values were found in samples taken after the examination compared with before, in all dogs and within all breeds. No breed differences were found within the sample taken before the examination, whereas ratios in the sample taken after the examination were higher in both CKCSs and Dachshunds compared with Labrador Retrievers (P < .0001 for both), Table 3. The increase in epinephrine/creatinine ratio was significantly greater in both CKCSs and Dachshunds compared with Labradors (overall P < .0001, Tukey P < .05).
Comparing norepinephrine/creatinine ratios, significantly higher values were found in samples taken after the examination compared with before, in all dogs, including CKCSs and Labrador Retrievers, while in Dachshunds ratios did not differ between samples taken before and after the examination. Labrador Retrievers had lower ratio in the sample taken before the examination compared with CKCSs (P = .020) and Dachshunds (P = .013), as well as in the sample taken after the examination (P = .0008 compared with CKCSs and P = .035 compared with Dachshunds). The increase in norepinephrine/creatinine ratio was significantly greater in CKCSs compared with Labrador Retrievers (overall P = .04, Tukey P < .05), Table 3.
Epinephrine/creatinine and norepinephrine/creatinine ratios in the 13 Labrador Retrievers examined both at the clinic and at home are shown in Table 4. In the sample taken after the examination, epinephrine/creatinine ratio was significantly higher when the dog had been examined at the clinic than in the home setting (P = .005), whereas norepinephrine/creatinine ratios did not differ.
|Epinephrine/Creatinine Ratio (nmol/mmol)||Adj P-Value||Norepinephrine/Creatinine Ratio (nmol/mmol)||Adj P-Value|
|Examination at clinic|| |
1.5 ± 0.6
5.0 ± 2.5
4.6 ± 2.9
6.6 ± 3.0
|Examination at home|| |
1.7 ± 1.0
2.8 ± 1.1
3.2 ± 1.7
6.4 ± 4.4
There were weak, but positive, associations between epinephrine/creatinine ratios in the urine sample taken after the examination (n = 56), and SBP (R2 = 0.11, P = .012), DBP (R2 = 0.11, P = .010), and HR (R2 = 0.16, P = .002) in the standard setting, as well as norepinephrine/creatinine ratios after examination (n = 56), and SBP (R2 = 0.17, P = .002), and DBP (R2 = 0.15, P = .003), but not HR (R2 = 0.13, P = .067), in the standard setting.
Many external and intrinsic factors may activate the sympathetic nervous system and affect BP, making interpretation of BP recordings difficult. In the present study, SBP and DBP increased when owners left dogs in examination room and recordings were made by veterinarian alone. Also HR increased when veterinarian was alone with the dog, alerting it with a squeaky toy. The positive association between HR and SBP and DBP suggests a stress-induced reaction, because the baroreceptor reflex should decrease HR when BP increases, unless other influences on HR are present. Both epinephrine/creatinine and norepinephrine/creatinine ratios increased in the sample taken after the examination compared with the sample taken before, confirming that dogs experienced stress or excitement during travel to and examinations at the clinic.
There are breed differences in BP with lower values in Labrador Retrievers. In the present study, DBP was lower in Labrador Retrievers than Dachshunds in both standard and owner settings. All BP recordings were performed on standing dogs using tail cuff.[25-27] As Labrador Retriever is a medium-sized breed and Dachshund a small-sized breed, a potential reason for lower BP values could be a larger difference in height between heart and tail in Labrador Retrievers, leading to underestimation of BP values. The actual differences were, however, larger than would have been expected from difference in height alone. Furthermore, no differences were seen between Labrador Retrievers and the small-sized CKCS, neither in SBP nor in DBP and because significant differences between Labrador Retrievers and Dachshunds were found only in DBP and not in SBP, a systematic underestimation of BP in Labrador Retrievers is unlikely.
In all 3 breeds, mean values of both SBP and DBP were within reference values for healthy dogs examined by oscillometric devices. The majority of dogs had SBP values < 160–170 mmHg, a cut-off point typically used for diagnosis of canine hypertension. However, SBP of some individuals reached up to 180 mmHg in standard setting and up to 200 mmHg when measured by veterinarian alone. Also DBP exceeded the reference values in some dogs. The thorough general health examination ascertained that all included dogs were healthy and therefore the high BP values in certain dogs are interpreted as stress-induced. In support of this interpretation is the effect of examinations on the urinary catecholamine concentrations.
Both epinephrine/creatinine and norepinephrine/creatinine ratios increased after examination. Similar findings were made in a study comparing samples obtained after clinical examination to samples taken at home after adaptation to sampling procedure. However, in the present study, higher ratios of both epinephrine/creatinine and norepinephrine/creatinine after the examination as well as norepinephrine/creatinine ratio before the examination were seen in both Dachshunds and CKCS compared with Labrador Retrievers. This could potentially be caused by a different reactivity pattern in the sympathetic nervous system. For that suggestion speaks the higher DBP in Dachshunds compared with Labrador Retrievers in the first 2 settings and the higher HR in CKCSs compared with Labrador Retrievers in the first 3 settings. The differences between breeds disappeared when the Labradors reacted with increased DBP and HR in the last setting. Labrador Retrievers are known to be calm and are often used as a guide dog for the blind.
The CKCSs had higher HR than Labrador Retrievers in all settings except the squeaky toy, while Dachshunds did not differ from the other breeds. In contradiction to the common assumption that small dog breeds have higher HRs than large breeds,[29, 30] several studies have not found a correlation between HR and body size.[4-6, 31] Breed type, on the other hand, has been associated with HR and spaniels have been shown to have higher HRs than several other breed types. In a recent study of healthy dogs, CKCSs were shown to have higher minimum and mean HR on 24-hour Holter recordings compared with Dachshunds.
Conflicting results exist from previous studies comparing BP and HR recordings at home to clinical settings.[27, 33-35] Among the 13 Labrador Retrievers in the present study, no differences were found in BP or HR between examinations at clinic and at home, but epinephrine/creatinine ratio was lower after examination at home compared with clinic, suggesting that home measurement may have been pernceived less stressful by dogs.
In the present study, a HDO device was used for BP and HR recordings. Two recent studies have evaluated the HDO machine[18, 19] against radiotelemetry, which has emerged as the best invasive method to measure blood pressure in conscious animals. In 1 study, measuring BP and HR at baseline and after administration of a drug that increases BP, both telemetry and HDO were found highly reproducible with high intra-individual precision for SBP, DBP, and HR, although HDO showed slightly higher standard deviation for all variables. In the 2nd study, comparing telemetry, HDO and Cardell BP measurements in conscious dogs at baseline and after administration of a BP-lowering drug, moderate-to-strong correlation for BP and very strong correlation for HR were found between HDO and telemetry. In both studies, the ranges of BP obtained during experiments were similar to ranges obtained in the present study. In the 1st study, HDO was found to overestimate BP values compared with telemetry, while the 2nd study found BP values to be underestimated by HDO. The differences could be caused by the HDO machine, but the radiotelemetry system also gives different baseline values. Once the system is implanted in the animal, no calibration can be made and there is a drift in the system. Both BP and HR change within seconds in dogs. Hence, variability between subsequent measurements may not only be caused by variations in the BP device used, but also by intradog variations during the recording.
In the present study, recordings were performed by the same trained and experienced veterinarian. No exclusion of readings was made during recordings. Recordings were also made by dog owners. In a previous study, examiners with different levels of training measured BP using HDO and Doppler techniques, however without comparison with direct BP measurements. Irrespective of observer experience, all within- and between-day CVs for SBP were < 15% and all attempts of both SBP and DBP recordings were successful using HDO in that study, even for the least experienced examiner. Also in our study, HDO recordings were easily obtainable by most dog owners after a short instruction. The BP measurements were performed in the same order in all dogs. The standard setting with both owner and veterinarian present was performed first, for owners to get introduced to the protocol and to the measurement procedure. As a carry-over effect from one setting to the next was expected, settings were thereafter performed starting with the setting assumed to be least stressful for the dog (owner alone) and ending with the setting assumed most stressful for the dog (squeaky toy). Based on the results with lower values of SBP, DBP and HR in the standard setting compared with one or both settings without owner present, we recommend owners be present when performing BP recordings.
A diurnal variation in BP has been demonstrated in several species, including humans, rats, cats, and dogs.[39-42] Also, norepinephrine concentrations have been shown to vary diurnally. Examinations of the present study were therefore performed at a standardized time in the morning. Sodium intake is known to affect BP in humans. Even though there is no strong evidence that dietary sodium affects BP in dogs or cats, efforts were made to avoid excessive salt intake in the present study to avoid a potential confounding factor.
This study validates ELISA assays for analysis of catecholamine concentrations in canine urine. In a study comparing ELISA with HPLC for measurement of catecholamine concentrations in human plasma and urine, the ELISA was found accurate, sensitive, specific, and precise. Analysis of epinephrine and norepinephrine in canine urine by HPLC in 3 recent studies[10-12] gave comparable values in healthy dogs to those obtained in the present study. The in-house validation showed acceptable performance of epinephrine and norepinephrine ELISA assays for canine urine, although interassay CVs for both assays were relatively high. Stability in the freezing step for canine urinary catecholamines has been tested. As in the case for human urine,[8, 9] results showed that compounds were stable in canine urine over time, The ELISA method is less expensive and time consuming compared with HPLC and could therefore be a good alternative.
As adaptation to sampling procedure has been shown to lower catecholamine concentrations in dogs, owners were instructed to practice urine sampling at home at least 3 times during the fortnight preceding the examination. However, with this protocol, it was not practically possible to collect urine after each of the 4 clinical settings, which would have been interesting to directly relate BP and HR values to catecholamine concentrations. Instead, catecholamine concentrations in the sample taken after examination represent summation of stress or excitement experienced by dogs during travel to the clinic as well as during the examinations taking place in the clinical environment. To investigate effects of travel and clinical environment compared with the examinations themselves, home examinations were included in the research plan. For practical reasons, home examinations were performed only in a small number of dogs and results therefore need to be interpreted cautiously. BP and HR did not differ significantly between clinic and home and as epinephrine/creatinine ratios were higher after examination both at home and at the clinic compared with before, the examinations themselves should be perceived as stressful by the dogs.
This study has limitations. Blood pressure values were not adjusted for hydrostatic pressure differences, which could have led to an underestimation of BP in Labrador Retrievers, having a longer distance from tail to base of heart compared with the other 2 breeds. Dilutional linearity was not performed for catecholamine ELISA assays during validation, as spike-and-recovery was considered acceptable. However, as dilutional linearity might have added further information, the lack of this assessment is a potential study limitation. Furthermore, accuracy testing was not performed, as validation was performed for a research situation. To validate the assays for clinical use by a commercial laboratory, further studies would be needed including comparison with the gold standard HPLC. Such a comparative study was beyond the scope of the present study. The present study only included dogs of 3 breeds. Only male dogs were included and they were all relatively young. The study population can therefore not be seen as representative of the general canine population. Also, the statistically significant differences between breeds representing physiological variations seen in the present study are small, and might thus not be relevant during clinical examination of an individual dog.
In conclusion, BP and HR were higher when recordings were made by veterinarian alone. We therefore recommend owners be present when performing BP recordings. The potential stress or excitement induced by travel and examinations at the veterinary clinic was confirmed by significantly higher catecholamine/creatinine ratios in urine samples taken after the examination compared with samples taken before the examination. A notable observation was very high SBP and DBP readings in a few individual dogs. Several breed differences were found, with Labrador Retrievers having low DBP, HR, and catecholamine concentrations, possibly caused by less mobilization of the sympathetic nervous system in this breed. Further studies on breed differences are warranted.
The skillful technical assistance of BMA Gunilla Drugge is greatly appreciated. The study was supported by the European Commission (FP7-LUPA, GA-201370), Agria Insurance Company's Research Foundation, Sweden, and Djurskyddet, Sweden.
Conflict of Interest: Authors disclose no conflict of interest.
Vet HDO Memodiagnostic SN1908, S + B medVet Systeme Beratung, Germany
Cardio Perfect, Welch Allyn AB, Svärdvägen 21, 182 33 Danderyd, Sweden
Philips HD11XE, Philips, Bothell, Seattle, WA
IBL International Gmbh, Hamburg, Germany
Assay Designs Inc, Ann Arbor, MI
SAS 9.2, SAS Institute Inc, Cary, NC