Influence of clinical setting and cat characteristics on indirectly measured blood pressure and pulse rate in healthy Birman, Norwegian Forest, and Domestic Shorthair cats

Abstract

Background

Measured indirect blood pressure (BP) results in cats in a clinical environment might be affected by stress and characteristics of the cats.

Hypothesis

To investigate the influence of clinical setting, cat characteristics, and life situation on BP and pulse rate (PR) in healthy cats.

Animals

Ninety‐four healthy Domestic Shorthair, Birman and Norwegian Forest cats.

Methods

Blood pressure measured by high‐definition oscillometry in 3 settings: cat placed in its own carrier with veterinarian present; cat placed in carrier with owner alone present; and cat placed on table with veterinarian present. Statistical analyses were performed using mixed linear models.

Results

Systolic BP (SBP) did not differ among settings. Higher mean arterial pressure (MAP), diastolic BP (DBP), and PR were found when measurements were performed with cat placed on table, rather than in carrier. Coefficients of variation (CVs) higher for SBP, MAP, DBP, and PR when measured with cat placed on table than in carrier. Birman cats had lower BP than other breeds. Systolic BP, MAP, DBP, and PR increased with age. Cats allowed outdoors had lower PR than cats living strictly indoors.

Conclusion and Clinical Importance

No difference in SBP was found among settings, but measuring BP with the cat placed on the examination table gave higher MAP, DBP, PR, and CV than measuring BP with the cat in its carrier. Breed affected BP, with lower BP in Birman cats than other breeds. Blood pressure increased with age. Pulse rate was lower in cats allowed outdoors than cats living strictly indoors.

Abbreviations

APPW

arterial pulse pressure wave

BCS

body condition score

BP

blood pressure

BW

body weight

Carrier‐O

carrier‐owner

Carrier‐VO

carrier‐veterinarian‐owner

CI

confidence interval

CV

coefficient of variation

DBP

diastolic blood pressure

DSH

non‐purebred Domestic Shorthair

FS

fractional shortening

HDO

high‐definition oscillometry

HR

heart rate

IVSd

interventricular septum diastole

IVSd_inc%

percentage increase interventricular septum diastole

LA

left atrium

LA/Ao

left atrial‐to‐aortic root diameter ratio

LV

left ventricular

LVFWd

left ventricular free wall diastole

LVFWd_inc%

percentage increase left ventricular free wall diastole

LVIDd

left ventricular internal diameter diastole

LVIDd_inc%

percentage increase left ventricular internal diameter diastole

LVIDs

left ventricular internal diameter systole

MAP

mean arterial blood pressure

NF

Norwegian Forest

PR

pulse rate

SBP

systolic blood pressure

Table‐VO

table‐veterinarian‐owner

TOD

target organ damage

TT4

total thyroxine

1. INTRODUCTION

Regulation of blood pressure (BP) is complex, involving the cardiovascular, nervous, renal, and endocrine systems. Excitement, stress, and anxiety in a clinical environment might increase BP by sympathetic activation (ie, situational hypertension ¹ ) in cats, ² dogs, ³ , ⁴ and humans, ⁵ impairing the reliability of indirect BP measurements. The clinical setting has been shown to affect BP measurements in dogs, with higher systolic BP (SBP) and diastolic BP (DBP) results when performed by the veterinarian alone, compared to when the owner was present. ³ Few published studies have investigated how different clinical settings might affect measurements of BP and pulse rate (PR) in healthy cats. ⁶ , ⁷

Breed differences have been identified for BP and PR in dogs, ³ , ⁸ whereas to the authors' knowledge, no studies have been specifically designed to investigate breed differences for these parameters in cats. Previous studies on other cat characteristics and their association with BP and PR measurements in healthy cats show variable results. ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶

In the clinical environment, BP measurements in cats are performed routinely by oscillometric or Doppler methods. ¹⁰ , ¹¹ , ¹³ , ¹⁵ , ¹⁷ Indications for BP measurement in cats might be to identify and monitor systemic hypertension and hypotension in cats with primary disease processes that might affect BP, and potentially to screen for idiopathic hypertension in older cats. ¹ , ¹⁸ Measurement of BP is indicated when evidence of target organ damage (TOD), such as hypertensive ocular TOD and left ventricular (LV) hypertrophy, is present, and hypertension thus might be suspected. ¹ , ¹⁹

Our primary aims were to investigate how different clinical settings (cat in a carrier with the owner and veterinarian present, cat in carrier with only the owner present, cat on the table with the owner and veterinarian present) and cat characteristics (breed, age, sex, body weight, and body condition) affect indirectly measured BP and PR in healthy cats. A secondary aim was to evaluate the potential influence of life situation (eg, single cat household, multi‐cat household, being allowed outdoors, living strictly indoors) on BP and PR in healthy cats. We hypothesized that indirect BP and PR results would be affected by the clinical setting during measurement, cat characteristics, and life situation.

2. MATERIALS AND METHODS

This prospective observational randomized study was approved by Uppsala Animal Experiment Ethics Board, Sweden (C137/13). Client‐owned cats were recruited for the study by information distributed to cat owners on webpages, at seminars for owners, or at the recruiting clinic. Cats were examined at Evidensia Animal Clinic in Västerås in Sweden between September 2014 and June 2017. Informed written consent was obtained from the owner of each cat. The population has in part been described in a previous study. ²⁰

2.1. Study design

Each cat was examined between 9:00 am and 1:00 pm by a Swedish national veterinary specialist in internal medicine and cardiology (SH), according to a standardized protocol. Cats were brought directly to a quiet examination room upon arrival to the clinic, case history was obtained, and the cat was gently removed from the carrier and weighed on a digital scale. After an acclimatization period, BP measurements were performed in different clinical settings, followed by physical examination including assessment of body condition score (BCS; 1‐9), ²¹ echocardiography, and blood sampling, in the same manner, and in the order described. All examinations were performed without sedation of the cats. Some cats had urinalysis performed.

2.2. Inclusion criteria

Healthy Domestic Shorthair (DSH), Birman, and Norwegian Forest (NF) cats, aged 1‐14 years, were included in the study. Cats were considered healthy if their physical examination did not identify any clinically relevant abnormalities, and if echocardiogram, hematology, and biochemistry results were within normal reference intervals. Where urine was available, urinalysis was performed and results were required to be within normal reference ranges for the cat to be included.

2.3. Exclusion criteria

If findings indicating clinically relevant organ‐related, or systemic diseases were detected, or if hematology or blood biochemistry results were outside reference intervals, the cat was excluded. Cats receiving medical treatment were excluded. If the BP protocol was not followed, the cat was excluded.

2.4. Indirect blood pressure measurement

For acclimatization, all cats were allowed to adapt to the clinical environment for at least 10 to 15 minutes in the presence of their owners (ie, allowed to roam freely) before BP measurement was performed. ¹ , ¹⁸ Indirect BP was measured using high‐definition oscillometry (HDO) device (Vet Memodiagnostic HDO monitor, S+B medVET, Babenhausen, Germany) with the C1 cuff applied to the base of the tail. Measurements were taken in 3 different clinical settings: 2 with the cat placed in its own carrier (a covered pet carrier) and 1 with the cat placed on the examination table (with rubber matting). The clinical settings were as follows: (a) cat placed in its carrier with the owner and veterinarian present in examination room. The veterinarian performed 6 measurements (setting carrier‐veterinarian‐owner [Carrier‐VO]); (b) cat placed in its carrier and after instructing the owner on use of the device, the veterinarian left the room, and the owner performed 6 measurements (setting carrier‐owner [Carrier‐O]); and (c) cat placed on the examination table, with both the owner and veterinarian present. The veterinarian performed 4 measurements (table‐veterinarian‐owner [Table‐VO]).

At least 10 seconds were allowed to elapse between all cuff inflations. The sequence of settings was randomized for all cats using Microsoft Excel. For practical reasons, the 2 settings with the cat placed in the carrier were always carried out sequentially, but in randomized order, whereas the setting with the cat placed on the table was randomized to be performed either first or last. In settings Carrier‐VO and Carrier‐O, the cuff was placed on the tail after the acclimatization period, and the cat was gently placed in the carrier, which then was closed. The cat had the cuff on the tail for 5 minutes before the measurement started. In setting Table‐VO, the cat was gently placed on the table after the acclimatization period, the cuff was placed on the tail and the cat was gently held by the owner during measurements. The cat was allowed to settle and when it was still on the table, standing or resting in sternal recumbency, BP measurement started. For all settings, SBP, MAP, DBP, and PR were recorded using the HDO device. No readings were excluded at the time of measurement. If the HDO device failed to obtain a reading and a new cuff inflation had to be performed, it was noted as an error by the veterinarian in settings Carrier‐VO and Table‐VO. Errors could not be noted by the veterinarian in the setting Carrier‐O, with the owner only present.

In some cats, subjective evaluation of the arterial pulse pressure wave (APPW) form was performed by the veterinarian after each measurement in settings Carrier‐VO and Table‐VO. The APPW form was assessed as “adequate” if the pulse waves generated a bell curve pattern, and “inadequate” if the waveform had substantial distortion (Supplement 1). ²²

2.5. Echocardiography

Echocardiographic examination was performed using an ultrasound unit (IE33, Philips Ultrasound, Bothell, Washington) with a 4 to 12 MHz phased‐array probe, and continuous ECG monitoring. ²³ Left atrial‐to‐aortic root diameter ratio (LA/Ao) was measured from the right 2‐dimensional (2D) short‐axis view. ²⁴ End‐diastolic and systolic LV dimensions (interventricular septum diastole [IVSd], LV internal diameter in diastole [LVIDd], LV free wall diastole [LVFWd], LV internal diameter in systole [LVIDs], and fractional shortening [FS]) were measured from M‐mode and 2D images. ²³ , ²⁵ Expected body weight (BW)‐dependent values for IVSd, LVIDd, and LVFWd were calculated according to previously generated formulas for cats. ²⁶ Mitral, tricuspid, aortic, and pulmonic valves were interrogated using spectral and color flow Doppler. ²³ , ²⁵

2.6. Blood and urine analyses for health assessment

Blood sampling was performed by venipuncture for analyses of selected hematology (hematocrit, hemoglobin, white blood cell count), and biochemistry tests (alanine aminotransferase [ALT] activity, and serum creatinine, glucose, total protein concentrations) at Evidensia Animal Clinic in Västerås using the ProCyte (IDEXX ProCyte Dx, IDEXX Laboratories, Inc., Westbrook, Maine) and Catalyst (Catalyst Dx Chemistry Analyzer, IDEXX Laboratories, Inc., Westbrook, Maine) systems. Serum samples for total thyroxine (TT4) and fructosamine concentrations were analyzed by chemiluminescence at the Clinical Pathology Laboratory at the University Animal Hospital of the Swedish University of Agricultural Sciences, using the Immulite (IMMULITE 2000, Siemens Healthcare GmbH, Erlangen, Germany) and Abbott Architect (Abbott Architect c4000, Abbott Park, Illinois) systems, respectively.

A voided urine sample from the cat was collected by the owner at home, if possible. Urine samples were examined by urine dipstick analysis (Siemens Multistix 10SG, Erlangen, Germany). Urine specific gravity was measured using a digital refractometer (Pocket refractometer, Atago, Tokyo, Japan).

2.7. Statistical analyses

Statistical analyses were performed using commercially available software (SAS 2017, SAS Institute Inc., Cary, North Carolina), ²⁷ (JMP Pro 12.2.0, SAS Institute Inc., Cary, North Carolina). Data on SBP, MAP, DBP, and PR are presented as mean ± SD. Coefficient of variation (CV) is presented as mean in percentage. Group comparisons for cat characteristics and echocardiographic variables were made using 1‐way analysis of variance (ANOVA). Significance level was set at P < .05 and multiple comparisons were adjusted using Tukey's method.

When analyzing the effect of clinical setting on obtained indirect BP and PR results, readings for the first 4 measurements per setting initially were included to obtain a balanced data set (ie, the 2 last measurements for Carrier‐VO and Carrier‐O were excluded). In the next step, measurements with PR < 75 or > 300 pulses/min, regarded as outliers, were excluded. ²⁸ , ²⁹ , ³⁰ Mean values, SD, and CV for SBP, MAP, DBP, and PR were determined for each cat in each setting. These summary statistics were subjected to analysis without transformation by mixed linear models using the mixed procedure in SAS. ²⁷ , ³¹ Models included setting, sequence, and the interaction setting × sequence as fixed factors. Cat was set as random factor. If significant interactions were found, subsequent comparisons of least squares mean values were performed to investigate in which sequences the BP, PR, and CVs differed. Assumptions underlying analysis were checked by preparing diagnostic plots. Post hoc comparisons were adjusted for multiplicity using Tukey's method. Although sample estimates of SD and CV are not expected to follow normal distributions, sample size was large enough to allow analyses using similar models as for mean values, and residual plots did not suggest extreme distributions.

Randomization of BP settings gave the following possible sequences: (Carrier‐VO; Carrier‐O; Table‐VO); (Carrier‐O; Carrier‐VO; Table‐VO); (Table‐VO; Carrier‐VO; Carrier‐O); and (Table‐VO; Carrier‐O; Carrier‐VO). Each of the 94 cats was subjected to each of the 3 clinical settings, resulting in 282 sessions.

Because of the high CVs for BP and PR measurements in setting Table‐VO, only settings Carrier‐VO and Carrier‐O were included in evaluation of the effect of cat characteristics and life situation on measured BP and PR. Six measurements per setting were available. After discarding the first BP measurement in settings Carrier‐VO and Carrier‐O, measurements with >20% variation in SBP ¹ , ¹⁸ , ²² and measurements with PR < 75 or > 300 pulses/min, ²⁸ , ²⁹ , ³⁰ 1 mean value for SBP, MAP, DBP, and PR, respectively, was determined for each cat. Models then were built using stepwise procedures for general linear models, using the GLMselect procedure. ²⁷ The models included the following variables and all 2‐way interactions between them: breed, age, sex, BW, BCS, and life situation (ie, indoor only or allowed outdoor, number of cats in household). Age, BW, and number of cats in household were assessed as continuous variables. Body condition score was divided into 4 classes (4–7), sex into 4 classes (male, female, neutered male, neutered female), and breed was set as a classification variable. The first analyses were performed using the Lasso method, ³² because comparisons based on simulations indicated that the prediction power was slightly better for Lasso than for traditional methods, such as stepwise selection. ³³ These analyses suggested that breed and age were associated with all BP variables and PR. Therefore, new stepwise analyses for SBP, MAP, DBP, and PR in which breed and age were forced into the model were performed to determine if any other variables contributed. The assumptions underlying the analyses were checked using diagnostic plots. Post hoc pairwise comparisons were adjusted for multiplicity using Tukey's method.

3. RESULTS

3.1. Study population

Of 117 examined healthy cats, 23 were excluded. Five cats had kidney disease, 5 heart disease, 2 severe gingivitis, 2 congenital defects (diaphragmatic hernia and peritoneal‐pericardial diaphragmatic hernia), 2 increased ALT activity, 1 hyperthyroidism, 2 treated with deslorelinacetate, and in 4 cats BP was not measured according to study protocol. Blood glucose concentration was mildly increased in 6 cats. However, serum fructosamine concentration, hematology, and other serum biochemistry variables were within normal limits, and thus these cats were included. Urine was available in 38 of 94 cats and results of urinalyses were within normal limits in all cats (Table 1).

TABLE 1.

Cat characteristics, echocardiographic data, and laboratory variables in 94 healthy cats

Group	All cats	Birman	Domestic Shorthair	Norwegian Forest	P‐value
Number	94	34	27	33
Sex (F/M)	54/40	21/13	12/15	21/12
Neuter status (F/NF/M/NM)	22/32/9/31	12/9/6/7	0/12/0/15	10/11/3/9
Indoor/Outdoor	68/26	30/4	9/18	29/4
Age (y)	5.4 ± 3.8	4.7 ± 3.9a	6.7 ± 4.3a	5.0 ± 3.0a	.11
BCS normal 4–5/ Overweight 6–7	48/46	23/11	11/16	14/19
BW (kg)	4.6 ± 1.4	3.6 ± 0.7a	4.8 ± 1.0b	5.4 ± 1.6b	.0003
HR ausc (bpm)	158 ± 26	158 ± 20a	151 ± 28a	165 ± 27a	.11
LA/Ao	1.1 ± 0.1	1.1 ± 0.1a	1.1 ± 0.1a	1.1 ± 0.1a	.68
IVSd (mm)	3.8 ± 0.4	3.6 ± 0.4a	3.9 ± 0.4b	4.0 ± 0.4b	.0032
IVSdinc (%)	−0.5 ± 9.1	−1.1 ± 9.4a	0.4 ± 8.7a	−0.6 ± 9.4a	.82
LVIDd (mm)	16.2 ± 2.1	15.1 ± 1.6a	16.5 ± 1.8b	17.1 ± 2.3b	.011
LVIDdinc (%)	2.5 ± 10.3	1.1 ± 9.0a	3.4 ± 12.4a	3.3 ± 10.0a	.62
LVFWd (mm)	3.7 ± 0.5	3.5 ± 0.4a	3.8 ± 0.5b	3.9 ± 0.4b	.015
LVFWdinc (%)	−0.9 ± 9.6	−2.0 ± 9.5a	−0.2 ± 10.7a	−0.5 ± 8.8a	.74
FS (%)	50 ± 7	49 ± 7a	52 ± 8a	50 ± 7a	.15
Creatinine (mg/dL)	1.68 ± 0.3	1.80 ± 0.32a	1.58 ± 0.27b	1.64 ± 0.26a,b	.011
Hematocrit (%)	35 ± 5	35 ± 7a	35 ± 4a	37 ± 4a	.28
TT4 μg/dL (nmol/L)	2.4 ± 0.5 (30.6 ± 7.0)	2.3 ± 0.6a (28.9 ± 7.3)	2.7 ± 0.6b (34.3 ± 7.4)	2.3 ± 0.4a (29.4 ± 5.2)	.016
USG	1.053 ± 0.01 (N = 38)	1.057 ± 0.01a (N = 19)	1.053 ± 0.01a (N = 9)	1.047 ± 0.01a (N = 10)	.10

Note: The mean ± SD are shown for continuous variables. Within each row, values with different superscripts are statistically different between breeds. Tukey's adjustment for multiple comparisons was performed and a P‐value of <.05 was considered significant. For urine‐specific gravity (USG), results were available for a proportion of cats, and the number of cats is stated in the table.

Abbreviations: BCS, body condition score; BW, body weight; F, female; FS, fractional shortening; Indoor, indoor only; HR ausc, heart rate auscultation; IVSd, interventricular septum diastole; IVSd_inc, percentage increase interventricular septum diastole; LA/Ao, left atrial‐to‐aortic root diameter ratio; LVFWd, left ventricular free wall diastole; LVFWd_inc, percentage increase left ventricular free wall diastole; LVIDd, left ventricular internal diameter diastole; LVIDd_inc, percentage increase left ventricular internal diameter diastole; M, male; N, number of cats for which results were available; NF, neutered female; NM, neutered male; Outdoor, allowed outdoors; TT4, total thyroxine; USG, urine specific gravity.

Of the 94 included cats, 33 were NF, 34 Birman, and 27 non‐purebred DSH. Cat characteristics, echocardiographic, and laboratory variables are presented in Table 1.

3.2. Indirect blood pressure and pulse rate

In the data for analysis of effect of clinical setting on obtained BP and PR results, there were 12 missing BP values (11 in setting Table‐VO; cats did not cooperate, and 1 in Carrier‐O; owner missed 1 measurement). Because each of the 94 cats was subjected to each clinical setting, there were 1116 measurements. Of these, 18 measurements from 13 cats were identified as outliers and excluded (2 in setting Carrier‐VO, 4 in Carrier‐O, and 12 in Table‐VO). The HDO device failed to obtain a reading and a new cuff inflation had to be performed in 55/94 cats. These errors occurred 46 times in setting Carrier‐VO and 65 times in Table‐VO.

Systolic BP did not differ between settings (P = .21). Clinical setting had an impact on MAP, DBP, and PR (overall P = .003). Mean arterial BP, DBP, and PR were higher in setting Table‐VO than Carrier‐O (P = .002). Furthermore, DBP and PR were higher in setting Table‐VO than Carrier‐VO (P = .007). Sequence alone did not significantly affect BP variables or PR. The combined effect of setting and sequence was significant for SBP, MAP, and DBP (all P = .02). Subsequent comparisons among settings, within sequences, showed lower results for MAP in setting Carrier‐O compared to Table‐VO, when the owner measured BP last in the sequence (Table‐VO; Carrier‐VO; Carrier‐O; P = .02).

Mean results for SBP, MAP, DBP, and PR in different settings for the first 4 measurements are presented in Table 2. Plots of SBP, MAP, DBP, and HR in different settings by measurement number are presented in Figure 1A‐D.

TABLE 2.

Blood pressure and pulse rate values in the different clinical settings

	SBP	MAP	DBP	PR
Setting	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD
Carrier‐VO	138 ± 18a	94 ± 11a,b	70 ± 10a	152 ± 26a
Carrier‐O	136 ± 18a	93 ± 12a	70 ± 11a	151 ± 25a
Table‐VO	138 ± 19a	97 ± 14b	75 ± 13b	161 ± 32b
P‐value	.21	.002	.011	.007
All	137 ± 18	95 ± 13	72 ± 12	155 ± 28

Note: Obtained SBP, MAP, and DBP and PR measurements in 94 healthy cats in the different clinical settings. Values are given as mean ± SD. Blood pressure is measured in mmHg, and PR in pulses per minute. Obtained BP and PR values were included for the first four measurements of each setting (in total 12 missing values), and thereafter measurements with PR <75 or >300 pulses/min, regarded as outliers, were excluded (in total 18 measurements). Within each column, values with different superscripts were statistically different between settings. Tukey's adjustment for multiple comparisons was performed and a P‐value of <.05 was considered significant. The P‐values presented are the highest of the pairwise comparisons within each column.

Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements; Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; MAP, mean arterial blood pressure; mmHg, millimeters of mercury; mmHg, millimeters of mercury; PR, pulse rate; SBP, systolic blood pressure; Table‐VO = cat placed on examination table with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements.

FIGURE 1.

Arterial systolic (A), mean (B), and diastolic (C) indirect blood pressures, and pulse rate (D) in all cats (n = 94) in three different clinical settings by the number of measurement. Mean values are presented at each measurement. Obtained BP and PR values were included for the first four measurements of each setting. There were 12 missing values, one in the setting Carrier‐O in the second measurement, and the others in the setting Table‐VO in the second (N = 1), third (N = 3), and fourth (N = 7) measurement. Measurements with PR <75 or >300 pulses/min (N = 18) were regarded as outliers and excluded. Two of these were in the setting Carrier‐VO in the fourth measurement: in the setting Table‐Carrier‐O they were in the first (N = 3) and fourth (N = 1) measurements, and in the setting Table‐VO in the first (N = 4), second (N = 4), third (N = 1), and fourth (N = 3) measurements. Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements; Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; Table‐VO = cat placed on examination table with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements

3.3. Variation in obtained indirect blood pressure and pulse rate measurements

The CVs for SBP, MAP, DBP, and PR in the different settings are presented in Table 3. Clinical setting had an impact with higher CVs for SBP, MAP, DBP, and PR results in setting Table‐VO compared to both settings in the carrier (all P < .0001). Neither sequence nor the combined effect of setting and sequence affected the CV for BP values. The combined effect of setting and sequence affected the CV for PR results (P = .04). Subsequent comparison among settings, within sequences, showed lower CV for PR in settings Carrier‐O and Carrier‐VO compared to setting Table‐VO, when the owner measured BP first in sequence Carrier‐O; Carrier‐VO; Table‐VO; P = .02). The APPW was available for evaluation in 43/94 cats (Supplement 1, Figure S1A‐C). The percentage values of adequate and inadequate APPWs were 90% and 10%, respectively, in setting Carrier‐VO, and 62% and 38%, respectively, in setting Table‐VO.

TABLE 3.

Coefficients of variation for blood pressure variables and pulse rate

	Coefficient of variation (%)
Settings	SBP	MAP	DBP	PR
Carrier‐VO	7.3a	8.4a	12.8a	8.3a
Carrier‐O	9.1a	8.6a	12.2a	7.4a
Table‐VO	13.4b	14.3b	18.7b	12.7b

Note: Coefficient of variation (CV) for obtained SBP, MAP, and DBP variables and pulse rate (PR) in the three different clinical settings for the first four blood pressure measurements per setting in 94 healthy cats (in total 12 missing values, of these, one was in setting Carrier‐O, and 11 in setting Table‐VO). Measurements with PR <75 or >300 pulses/min, regarded as outliers, were excluded (in total 18 measurements, of these, two were in setting Carrier‐VO, four in setting Carrier‐O, and 12 in setting Table‐VO). The CVs are given as mean values in percent. Within each column, values with different superscripts were statistically different between clinical settings. Tukey's adjustment for multiple comparisons was performed and a P‐value of <.05 was considered significant.

Abbreviations: Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements; Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; DBP, diastolic blood pressure; MAP, mean arterial blood pressure; PR, pulse rate; SBP, systolic blood pressure; Table‐VO = cat placed on examination table with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements.

3.4. Effect of cat characteristics and life situation on indirect blood pressure and pulse rate measurements

Because each of the 94 cats was subjected to each carrier setting with 6 measurements per setting, and there was 1 missing result, there were 1127 measurements (Supplement Figure S2A‐D). After discarding the first measurement in each setting, there were 939 measurements. Of these, 88 measurements from 39 cats were identified as outliers and excluded, leaving 4 to 10 included measurements per cat. Mean BP and PR measurements for each breed are shown in Table 4. Ten cats had mean SBP > 160 mmHg, of which 6 were NF cats and 4 were DSH cats. Of these 10 cats, 9 of 10 cats had a BP recording of <150 mmHg in 1 of the settings during the study.

TABLE 4.

Blood pressure and pulse rate values divided per breed

Breed		SBP	MAP	DBP	PR
Domestic Shorthair	27	146 ± 14a (116‐180)	98 ± 10a (77‐123)	72 ± 9a (48‐97)	141 ± 28a (90‐215)
Birman	34	125 ± 12b (107‐159)	85 ± 9b (68‐106)	63 ± 8b (44‐84)	159 ± 22b (104‐193)
Norwegian Forest	33	141 ± 17a (109‐170)	96 ± 10a (78‐117)	72 ± 9a (58‐92)	152 ± 25a,b (97‐207)
P‐value		<.0001	<.0001	.001	.02
All cats	94	136 ± 17 (107‐180)	93 ± 11 (68‐123)	69 ± 10 (44‐97)	151 ± 26 (90‐215)

Note: SBP, MAP, and DBP blood pressure and PR values in 94 healthy cats. Values are given as mean ± SD and (range), and are based on the clinical settings Carrier‐VO and Carrier‐O. Outliers have been discarded. Blood pressure was measured in mmHg and PR in pulses per minute. Within each column, values with different superscripts are statistically different between breeds. Tukey's adjustment for multiple comparisons was performed and a P‐value of <.05 was considered significant. The P‐values presented are the highest of the pairwise comparisons within each column.

Abbreviations: Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements; Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; DBP, diastolic blood pressure; MAP, mean arterial blood pressure; PR, pulse rate; SBP, systolic blood pressure.

Results of multivariable analyses of BP and PR vs cat characteristics (breed, age, sex, BW, BCS) and life situation (indoor only, allowed outdoor, number of cats in household) are presented in Table 5. All indirect BP variables were lower in Birmans than in NF and DSH cats (SBP, MAP, DBP; adjusted R ² = 0.28, 0.26, 0.16, respectively). The PR was higher in Birman than DSH cats (adjusted R ² = 0.06). With increasing age, all indirect BP variables (SBP, MAP, DBP; adjusted R ² = 0.32, 0.32, 0.23; P = .01, .0003, .004, respectively), as well as PR (adjusted R ² = 0.10; P = .04) increased. Cats allowed outdoors had lower PR (adjusted R ² = 0.14; P = .02) as compared with cats living indoors only. Mean PR (pulses/min) ± SD was 157 ± 24 (= 68) for cats living strictly indoors, and 137 ± 26 (=26) for cats allowed outdoors.

TABLE 5.

Association between indirect blood pressure and pulse rate, and cat characteristics and life situation in 94 healthy cats

Mean	Variable	Estimate	P value	95% confidence interval
				Lower limit	Upper limit
SBP (mmHg)	Breed		<.0001
	DSH	3.839		−3	11
	Birman	−16.029		−23	−9
	NF	Baseline breed
	Age	1.006	.01	0.2	1.8
MAP (mmHg)	Breed		<.0001
	DSH	−0.044		−5	5
	Birman	−11.358		−16	−7
	NF	Baseline breed
	Age	0.769	.0031	0.3	1.3
DBP (mmHg)	Breed		.0001
	DSH	−1.948		−6	2
	Birman	−8.592		−13	−5
	NF	Baseline breed
	Age	0.687	.0039	0.2	1.1
PR (pulses/min)	Breed		.02
	DSH	−4.596		−19	10
	Birman	7.488		−4	19
	NF	Baseline breed
	Age	1.458	.036	0.1	2.8
	Life situation		.016
	Indoor	16.462		4	29
	Outdoor	Baseline life situation

Note: The effect of cat characteristics and life situation on indirect SBP, MAP, DBP, and PR using multivariable analysis in 94 healthy cats. A mean value from the measurements from settings Carrier‐VO and Carrier‐O was used for each variable. The final multiple regression model, including cat characteristics (breed, age, sex, BW, and BCS), and life situation (indoor only/allowed outdoor, and number of cats in household) had an adjusted R ² of 0.32 for SBP, 0.32 for MAP, 0.23 for DBP, and 0.14 for PR. In this table, variables that remained significant in the final model are reported. A P‐value of <.05 was considered significant.

Abbreviations: Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements; Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; DBP, diastolic blood pressure; DSH, Domestic Shorthair; Indoor, indoor only; MAP, mean arterial blood pressure; NF, Norwegian Forest; Outdoor, allowed outdoor; PR, pulse rate; SBP, systolic blood pressure.

4. DISCUSSION

In this group of healthy cats, MAP, DBP, and PR were higher when indirect BP was measured with the cat placed on the examination table (Table‐VO) compared to 1 or both settings in the cat's own carrier (Carrier‐VO and Carrier‐O). The CVs for SBP, MAP, and DBP as well as PR were higher in setting Table‐VO compared to both settings with the cat placed in the carrier. Breed and age had an effect on BP and PR variables, with lower BP in the Birman cats than in NF and non‐purebred DSH cats, and increasing BP and PR with increasing age.

The variation in recorded results was higher for all BP variables, as well as PR, when measurements were performed with the cat placed on the table compared to both settings with the cat in its own carrier (Table 3). In the first part of the study, our focus was to evaluate performance of the oscillometric system under clinical conditions in different clinical settings, and therefore the first measurement was not discarded. A higher percentage of inadequate APPWs in setting Table‐VO than in setting Carrier‐VO (Supplement 1, Figure S1A‐C) might contribute to the higher variation in setting Table‐VO. Moreover, the HDO device failed to obtain readings more often in the setting Table‐VO than in the setting Carrier‐VO and additional cuff inflations had to be performed, which could have had an impact on the stress of the cat. Inadequate APPWs with following high CVs could be attributed to movement or anxiety, which may have been exacerbated by failure to obtain readings and requirement for more cuff inflations during measurements on the table. ²² , ³⁴ , ³⁵ Overall, these factors indicate that results from setting Table‐VO were less reliable in our study. A potentially more stressful situation also might explain the higher results of MAP and DBP in setting Table‐VO compared to 1 or both settings in the carrier (Table 2). However, albeit statistically significant, differences in recorded values among settings were small and likely not of relevance to the individual cat in a clinical situation.

In contrast to MAP and DBP, no difference was found among settings for SBP. Although MAP and DBP showed a decreasing pattern by measurement number for both settings in the carrier, but not for setting Table‐VO, this difference in pattern was not as evident for SBP, which might explain the lack of difference among settings (Figure 1A‐C). Hence SBP, the most commonly assessed variable in clinical practice, ¹ did not differ among settings in our study of healthy cats. Because the degree of situational hypertension has been reported to be larger in cats with hypertension than in healthy cats, ² investigation of the effects of different clinical settings on BP recordings in cats with hypertension would be of interest.

For PR, recorded results were lower in both settings in the carrier compared to setting Table‐VO and a decreasing pattern similar to observations for MAP and DBP was evident in the carrier settings, but not in setting Table‐VO (Figures 1B‐D). The first measurement had a higher percentage of inadequate APPW, which might be caused by cuff inflation surprising the cat, leading to movement (Figure S1B‐ S1C). Again, situational stress might initiate early pressor and tachycardic responses, thereby causing early increases in BP and PR, which is in accordance with previous studies in cats ² and humans. ⁵ This observation emphasizes the importance of performing multiple measurements over a period of time and to exclude the first measurement as recommended. ¹ , ²

No significant effect of the sequence of measurements (created by randomization of settings) was found on BP measurements or PR. However, the interaction between setting and sequence was significant for SBP, MAP, and DBP, complicating interpretation of these findings. Subsequent evaluation showed lower MAP when the owner alone measured BP last with the cat in the carrier in sequence (Table‐VO, Carrier‐VO, Carrier‐O) than with the cat placed on table first. An explanation might be that placing the cat on the table, after the acclimatization period, entails a change in the cat's situation, leading to increased BP. ² , ⁷ The first set of measurements then was performed—a new event for the cat. The cat then was placed in the carrier, potentially perceived as a secure place, before further measurements were performed. Cats often respond to confrontation by hiding, and the carrier might be considered a safe place to hide for many cats. ³⁶ , ³⁷ Finally, the veterinarian, who together with the owner had been present in both of the above settings, an unfamiliar person to the cat, left the room and final measurements were performed by the owner, a familiar person to the cat. ³⁶ , ³⁷ The order of these changes for the cat in the 3 settings in this specific sequence might explain the lower MAP in setting Carrier‐O compared to Table‐VO. In the clinical environment, being placed in the cat's own carrier, with its familiar smell, might thus cause less stress than being placed on the table. ³⁶ , ³⁷

A significant interaction was found between setting and sequence for the CV of PR, with lower CV in both settings in the carrier than the setting on the table in sequence (Carrier‐O, Carrier‐VO, Table‐VO). An explanation might be that the first measurements were performed by the owner alone, a familiar person, and that the cat had the possibility to hide in the carrier, leading to less perceived stress in this setting. ³⁶ , ³⁷ Thereafter, the veterinarian entered the room, and further measurements were performed by the veterinarian (setting Carrier‐VO), but the cat remained in the carrier. The measurement procedure was already known to the cat, presumably inducing less anxiety for the cat in this setting. ² , ³⁶ , ³⁷ The third setting, Table‐VO, led to a change in the cat's situation (ie, being removed from carrier), which might lead to increased stress and the possibility for movement. ² , ⁷

The Birman breed had lower SBP, MAP, and DBP than NF and DSH cats (Table 4). In previous studies, no effect of breed was found for BP. ⁹ , ¹⁰ , ¹⁶ The Birman cats also had higher PR than DSH cats in our study. In dogs, breed differences have been reported for both BP and PR. ³ , ⁸ A previous study showed that women had comparatively lower BP and higher HR than men. ³⁸ Studies indicate that lower BP in women might be caused by shorter length of the arterial tree and smaller LV diameter, giving a lower stroke volume in comparison with men. ³⁸ , ³⁹ In our study, standard cuff size for cats (cuff C1) was used. In the multivariable model, including breed, BW and the 2‐way interaction between them, breed was associated with BP variables, whereas BW was not, indicating that the lower BP in the Birman cats was mainly a breed effect, and less dependent on smaller body size. However, breed and BW are highly covariate, and a smaller tail circumference in Birmans, relative to the cuff size, might have contributed to lower results in this breed. ¹⁵ , ¹⁸

In all 3 breeds, mean SBP was within previously published reference ranges for cats obtained using oscillometric devices, ¹⁰ , ¹³ , ¹⁷ whereas mean DBP was under or at the lower end of previously published reference ranges. ¹⁰ , ¹⁷ The HDO performs a real‐time analysis of the arterial oscillations to obtain pulse amplitudes and measures SBP, MAP, and DBP. ⁴⁰ The lowest mean DBP was found in Birmans. This breed has a relatively low risk of being overweight ⁴¹ and also had the lowest proportion of overweight cats in our study. In people, it is important to measure DBP in relation to simultaneous magnitude of SBP. Diastolic hypertension occurs predominantly with obesity, insulin resistance, and hyperlipidemia and might evolve into systolic‐diastolic hypertension, a potential increased risk for future diabetes. ⁴² Obesity is a risk factor for diabetes mellitus in cats and is an important health problem facing domestic cats. ⁴³ Future studies on DBP in overweight and obese cats would be of interest, and for such studies, our results using healthy cats could serve as a foundation.

In our study, including cats of a wide range of ages, all BP variables increased with increasing age, in accordance with previous studies in cats, ⁹ , ¹⁰ , ¹¹ , ¹² dogs, ⁸ and humans. ³⁸ However, this is not a consistent finding in the literature and some studies of cats have shown no effect of age on indirect BP measurements. ¹³ , ¹⁴ , ¹⁵ These variable results might be caused by different cat populations, use of different BP devices, and different handling techniques of cats during measurements. Kidney disease can be a cause for hypertension in both cats and humans. Alterations in renal vasculature, such as fibrointimal hyperplasia and hyperplastic arteriolosclerosis, are found both in healthy cats and in cats with chronic kidney disease. ⁴⁴ In people and in cats, ⁴⁵ aging leads to progressive stiffening of arteries caused by vascular remodeling with progressive intimal thickening, which can lead to an increase in SBP with age, ⁴⁶ although these changes were found to be indistinguishable from similar lesions of aging in normotensive patients. ⁴⁷ In 1 study, hypertensive cats were not more likely to be affected by fibrointimal hyperplasia than normotensive cats. ⁴⁴ The process of vascular aging in cats and people is associated with vascular remodeling and inflammatory processes, ⁴⁴ , ⁴⁵ , ⁴⁷ which might be an explanation for increasing BP with age. Pulse rate also increased with increasing age, which is in accordance with a previous study. ²⁸ Heart rate also has been shown to increase with age in dogs. ⁸ , ⁴⁸

Pulse rate was lower in cats allowed outdoors than in cats living strictly indoors, possibly explained by cats allowed outdoors being more physically fit or more comfortable with different situations than cats living strictly indoors. However, this association was weak and should be interpreted with caution.

4.1. Study limitations

Our study included healthy cats of 3 genetically distant breeds, ⁴⁹ with relatively large group sizes. All cats were specifically recruited for the study. Cats visiting veterinary clinics for BP measurements comprise a great variety of breeds, including healthy cats as well as those with extra‐cardiac diseases that might lead to hypertension. Therefore, our study population cannot be considered representative of the general population, and results cannot be extrapolated to a general cat population or to cats with hypertension.

Kidney disease is common in cats and might cause hypertension. ¹¹ , ¹⁴ , ¹⁷ For practical reasons, urine samples were only available in 38/94 cats. Although serum creatinine concentration and hematocrit were within reference ranges in all cats, subclinical kidney disease might have been present in some cats, causing increased BP results. ¹¹ , ¹⁸ In accordance with previous studies, Birman cats had serum creatinine concentrations in the upper end of normal reference ranges, which might have masked subclinical kidney disease. ⁵⁰ , ⁵¹ Of the 94 cats, 10 had mean SBP > 160 mmHg, which might have been caused by situational stress or subclinical kidney disease. None of these cats were Birmans.

To limit the number of measurements for the individual cat and minimize the handling of each cat, 6 measurements were performed in each setting in the carrier (Carrier‐VO, Carrier‐O) and 4 measurements in the setting on the table (Table‐VO). In the first part of the study, focusing on the performance of the oscillometric system under clinical conditions, measurements 5 and 6 in the carrier settings were excluded to obtain a balanced data set. The first measurement of each setting was included, again to evaluate the oscillometric device under clinical conditions. Thus, the protocol deviated from current American College of Veterinary Internal Medicine guidelines. ¹ Because a large CV might affect findings by masking results, setting Table‐VO was excluded in statistical analysis of effect of cat characteristics and life situation on SBP, MAP, DBP, and PR. However, exclusion of measurements might affect the data.

Inspection of the APPW form was not performed in all cats, which is a limitation, considering that if the APPW was deemed inadequate, the result from that measurement should have been excluded. ²² Thus artifacts, such as motion disturbances, that might have affected measurements could not be evaluated in these cats. ²² , ⁴⁰

Another potential limitation is the slight difference in time from cuff placement to measurement between carrier and table settings. In carrier settings, the cat had the cuff on the tail for 5 minutes in the carrier before measurements started, whereas on the table measurement started when the cat had settled and was still on the table. This procedure was chosen to avoid unnecessary handling of the cats on the table.

A few results in our study had weak associations, and these should be interpreted with caution.

5. CONCLUSIONS

In our study of healthy cats, CVs were higher for indirect BP variables, as well as PR, when measurements were performed with the cat on the table compared to in the carrier. The SBP did not differ among the 3 clinical settings, whereas MAP, DBP, and PR were higher when measurements were performed on the table compared with the cat in its own carrier. The differences in recorded results for MAP, DBP, and PR among settings were small and likely not of clinical relevance for the individual cat. There was a higher percentage of inadequate APPW forms (evaluated by the veterinarian in some of the cats), when BP was measured on the table compared to in the carrier. Thus, measuring BP with the cat placed in its own carrier, thereby giving the cat the possibility to hide, might result in less variable BP. Blood pressure increased with increasing age and was lower in Birman cats than in both other breeds. Pulse rate was lower in cats allowed outdoors than in cats living strictly indoors.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approved by the Uppsala ethical committee for animal research, Uppsala, Sweden C137/1.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Supporting information

Figure S1 Subjective evaluation of the arterial pulse pressure wave (APPW) form.

Subjective evaluation of the APPW, that is, a visual display of pulse waves detected during BP measurement, was performed in 43 of the 94 included cats. The APPW form was evaluated by the veterinarian in settings Carrier‐VO and Table‐VO after each measurement. An APPW form was evaluated as “adequate” if the pulse waves generated a pattern of a bell‐curve, and “inadequate” if the waveform had significant distortion. When analyzing the APPW form in the clinical settings Carrier‐VO and Table‐VO, all obtained measurements for the first four measurements per setting were included, in order to obtain a balanced dataset, that is, the two last measurements for Carrier‐VO were excluded. Furthermore, measurements with PR <75 or >300 pulses/min were regarded as outliers and were excluded. As each of the cats was subjected to two clinical settings, and there were 3 missing BP values, this resulted in a total of 341 measurements. Of these, 7 measurements were identified as outliers and were excluded. Results of the APPW form were evaluated as adequate in 255 BP measurements and as inadequate in 79 BP measurements.

(A) Percentage of BP measurements evaluated as adequate and inadequate, respectively, based on the APPW form in setting Carrier‐VO is shown to the left, and in setting Table‐VO to the right. The higher percentage of adequate APPW forms in setting Carrier‐VO compared to setting Table‐VO indicates more reliable measurements in the carrier.

(B) Percentage of BP measurements evaluated as adequate and inadequate, respectively, in the setting Carrier‐VO by number of measurement. In the setting Carrier‐VO the percentage of BP measurements evaluated as adequate by number of measurement was >90% from the second measurement and onwards, indicating that the measurements in the carrier were reliable.

(C) Percentage of BP measurements evaluated as adequate and inadequate, respectively, in the setting Table‐VO by number of measurement. In the setting Table‐VO the percentage of BP measurement by number evaluated as inadequate was still 26% by the fourth measurement indicating less reliable measurements on the table in the present study.

Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; Table‐VO = cat placed on examination table with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements.

Figure S2 Obtained arterial systolic (A), mean (B), and diastolic (C) indirect blood pressures, and pulse rate (D) in all cats (n = 94) in the two clinical settings with the cat placed in its own carrier (Carrier‐VO and Carrier‐O) for six consecutive indirect blood pressure measurements. Mean values after exclusion of outliers (PR <75 or >300 pulses/min) are presented at each measurement.

Carrier‐VO = cat placed in carrier with cuff on tail, both owner and veterinarian present, veterinarian performed the measurements; Carrier‐O = cat placed in carrier with cuff on tail, owner alone performed the measurements.

Hanås S, Holst BS, Ljungvall I, et al. Influence of clinical setting and cat characteristics on indirectly measured blood pressure and pulse rate in healthy Birman, Norwegian Forest, and Domestic Shorthair cats. J Vet Intern Med. 2021;35:801–811. 10.1111/jvim.16096