Effect of physiological and pharmacological stress on heart rate, blood pressure, and echocardiographic measurements in healthy Warmblood horses

1 INTRODUCTION

Echocardiography is the diagnostic method of choice to investigate cardiac structures and to determine or compare cardiac dimensions.^1-6 The assessment of cardiac dimensions during repeated exams has an important prognostic value as progressive cardiac dilatation suggests progressive cardiac disease.^{3, 7, 8} Prerequisites to minimize measurement variability are well-defined imaging guidelines and adequate operator training.⁶ Between-day intraobserver and interobserver variability should be taken into account when evaluating alterations in cardiac dimensions during longitudinal follow-up.^{1, 6} Generally, the interobserver measurement variability appears to be a more important factor than the intraobserver variability, which emphasizes that preferably the same operator should perform sequential recordings and measurements in the same patient to minimize test-retest variability.⁶ However, slight changes in heart rate (HR) or blood pressure (BP) might influence measurements or measurement repeatability in case of follow-up examinations, among other related and determinative factors, such as preload, afterload, contractility, hydration, or cardiovascular status of the patient.⁴ The general advice is to perform echocardiographic examinations in adult horses at physiological HR ranging from 28 to 48 beats per minute (bpm) in the absence of external stressors. However, HR can be higher because of disease or stress, especially at the start of the examination, and often decreases as the echocardiographic exam continues.^{9, 10}

Sedation of the horse is usually contraindicated for echocardiography because of drug effects on HR, rhythm, contractility, preload, and afterload.⁴ Administration of detomidine hydrochloride resulted in decreases in HR and cardiac output (CO) with a tendency to increases in left atrial and ventricular dimensions.^11-15 Cardiac dimensions and function were also altered during physiological (treadmill exercise, lunging exercise) or pharmacological stress testing with atropine (parasympatholytic drug), or dobutamine (sympathomimetic drug), with high HR above 80 bpm and important increases in BP.^16-23 Postexercise evaluation of cardiac variables and pharmacological stress testing demonstrated decreases in left ventricular dimensions with increases in interventricular septum and left ventricular free wall thicknesses as well as an increased left ventricular fractional shortening (FS).^{18-20, 23-27} However, the influence of slight changes within the normal physiological range of HR on echocardiographic variables of cardiac function has not been investigated yet. Heart rate and BP might vary during the echocardiographic exam as a result of stress because of an unknown environment. This effect might also be imitated by pharmacological challenge. N-butylscopolammonium bromide (NBB) is a routinely used anticholinergic drug with an antagonistic effect on muscarinic receptors of the parasympathetic nervous system.²⁸ NBB is commonly used for the treatment of abdominal pain and bronchoconstriction in horses and has important cardiovascular effects as transient tachycardia and hypertension.^28-30

Aim of this study was to compare echocardiographic variables including 2-dimensional (2D), M-mode, pulsed wave (PW), and PW tissue Doppler imaging (PW TDI) in healthy, trained Warmblood horses at different physiological heart rates and during transient mild tachycardia and hypertension after IV administration of NBB and metamizol (NBBM). We hypothesized that cardiac dimensions would be decreased at higher HR and BP.

2 MATERIALS AND METHODS

3 RESULTS

The study sample included 11 geldings and 9 mares. The mean age of the horses was 14 ± 5 years. Median body condition score was 5/9 (range, 4-6/9).³¹ Mean height at the withers was 165 ± 5 cm, mean body length was 193 ± 6 cm, and mean chest circumference was 183 ± 11 cm. Mean estimated BW was 544 ± 71 kg. HR was significantly lower at rest compared with physiological stress and pharmacological challenge (Table 1). Mean arterial BP was significantly higher during physiological stress compared with rest, whereas DAP, SAP as well as pulse pressure (PP) did not significantly increase. Diastolic arterial BP, MAP, and SAP increased significantly between rest and pharmacological challenge (Table 1).

Table 2 summarizes the between-examination measurement variability and the comparison of the echocardiographic variables at rest, physiological stress, and pharmacological challenge. The CV of measurement variability between the 2 scans at rest can be classified as low for most of the echocardiographic variables. Very-low to low-measurement variabilities were seen for 2D and M-mode variables with exception for RV internal diameter at end-systole (RVIDs_{M_ch}) (moderate variability). Higher measurement variabilities were seen for PWD variables and calculated variables in general.

During physiological stress at mildly increased HR below 50 bpm, indices of left atrial contractile function differed significantly compared with rest, with increases in active LA FAC and A_m (Figure 1), whereas LAAa decreased. It should be noted that R-LAAd_end showed a smaller change compared with R-LAAa. LV FAC increased significantly. No significant differences were observed for most other 2D left atrial, left ventricular, aortic and pulmonary dimensions, or calculated variables of cardiac function. No significant changes in left-sided aortic PWD variables were present (Table 2).

Pharmacological challenge with NBBM led to significant changes in 2D, M-mode, PWD, and PW TDI echocardiographic variables compared with rest. Left atrial and diastolic left and right ventricular dimensions decreased. An increase in active left atrial fractional area change, diastolic septal wall thickness, left ventricular relative, and free wall thickness were noted. Aortic and pulmonary dimensions increased, whereas LV SV and EF decreased. In addition, a decrease in PW TDI variable S_m was seen (Table 2). All horses included in this study tolerated the pharmacological challenge test very well, and no pathological dysrhythmias were observed after administration of NBBM. Figure 2 depicts scatter plots indicating individual data points for the echocardiographic variables LAAa, active LA FAC, A_m, and LV internal diameter at end-diastole (LVIDd_{M_ch}) during the different scans.

4 DISCUSSION

Overall, no clinically relevant differences were found in our study for most 2D, M-mode, and PWD echocardiographic variables in healthy Warmblood horses at slightly increased heart rates during stress (40-45 bpm) compared with rest (30-35 bpm). Small, statistically significant differences were found for some echocardiographic variables at slightly increased physiological heart rate, but the mean difference fell inside the RC value provided for the respective variable, indicating that the measurement difference might be more likely the result of measurement variation than a true response to physiological stress (Table 2). However, physiological stress did result in significant increases in measures of left atrial contractile function (active LA FAC, Am). This might reflect an increased left atrial pump function at physiologically higher HR because of an altered autonomic tone and ventricular filling pressures. The left and right atrium act as a reservoir for venous return during ventricular systole and contribute in an important way to cardiac preload. During early-to-mid ventricular diastole, blood stored in the atria is emptied into the ventricles, whereas active atrial contraction establishes the final ventricular volume during late ventricular diastole.^{5, 48-50} As HR increases, diastolic ventricular filling time and preload decreases.^{50, 51} Active left atrial FAC increased significantly in our study because of decreases in left atrial area at atrial contraction. This finding of an increased active LA FAC demonstrates an enhanced booster pump function of the LA at slightly increased physiological HR in horses.^{40, 48, 50} Significant increases in PW TDI A_m further support an increased atrial pump function at slightly increased heart rates. TDI has been suggested to be a promising tool for a more sensitive, quantitative, global, or focal assessment of myocardial function in horses compared with other echocardiographic techniques.³⁹ LA active pump function and LV compliance are the major determinants of the A_m velocity, whereas the peak early-to-late diastolic filling velocity ratio (E_m/A_m ratio) reflects the relative contribution of atrial pump function to ventricular filling.^{5, 8, 39, 46, 48} An excellent correlation between A_m velocity and atrial function as well as LA ejection fraction, force, and kinetic energy has been demonstrated in a large number of studies in human medicine.⁵⁰ In healthy human individuals and cats, an age- and HR-related increase in tissue Doppler A_m velocities have been described.^{50, 52, 53} These findings emphasize the importance of HR while evaluating atrial contractile function by PW TDI echocardiography.⁵⁰ This increased left atrial pump function at physiologically higher HR should be considered during the echocardiographic assessment of horses under hospital conditions as this situation should be considered as stressful event. It should be noticed that active LA FAC was calculated from area measurements before atrial contraction (LAAp) and at maximal atrial contraction (LAAa), as previously reported.^40-42 The left atrial area at atrial contraction refers to the truly smallest area of the LA during cardiac cycle, and atrial contraction acts as additional filling of the LV before atrioventricular valve closure. Left atrial FAC calculated in this way was shown to be useful to assess the risk of atrial fibrillation recurrence postconversion.^{41, 42} Left atrial FAC has also been calculated using different area measurements, namely the area before atrial contraction (named LAAa in those studies, instead of LAAp) and the area at mitral valve closure (named LAAmin in those studies).^{5, 54} The latter corresponds to the LAAd_end of our study, which represents the atrial area after atrial contraction, and is influenced by the PQ interval. During physiological stress, LAAa showed a larger decrease (from 46.4 ± 6.2 cm² to 41.0 ± 6.0 cm², P < .001) compared with LAAd_end (from 51.0 ± 7.9 cm² to 47.2 ± 6.9 cm², P = .03). During pharmacological challenge, LAAd_end showed no significant decrease, whereas LAAa was significantly decreased (P < .001). These results suggest that the measurement during true atrial contraction more correctly reflects atrial booster pump function.

Pharmacological challenge with NBBM, a parasympatholytic agent with anticholinergic and antimuscarinic properties, caused similar echocardiographic effects as classical pharmacological stress testing, with decreased left ventricular dimensions and stroke volume and increased relative ventricular wall and interventricular septum thicknesses.^{18, 20, 23, 27} Cardiovascular effects of NBBM on HR and BP were visible within 2 minutes after IV administration of NBBM with a gradual reduction to baseline which occurred beyond image acquisition, similar to the findings of Morton et al,²⁸ who showed significant peak effects on HR and BP within 2 minutes after IV administration of NBB. Peak HR elevation occurred in the study of Morton et al 5 minutes post administration of NBB and persisted 25 minutes for DAP, MAP, and SAP and 50 minutes for HR with a gradual return to baseline.²⁸ Geimer et al documented a significant increase in HR and CO for at least 16 minutes and a short decrease in SV after administration of NBBM.⁵⁵ Pharmacological challenge in our study with NBBM led to significant decreases for most atrial and ventricular dimensions, volumes, and areas. Decreased left ventricular end-diastolic dimensions (LVIDd_{M_ch} and LV end-diastolic volume [LVEDV_al]) are considered as a surrogate for diastolic ventricular filling and a noninvasive measurement of cardiac preload (Table 2).^{51, 56} Although LA FAC was increased, LV A_m and E_m/A_m ratio were not altered. Therefore, it appears that the effects of an increased heart rate and a decreased diastolic LV filling time outweigh the effects of an enhanced left atrial booster pump function, leading to a decreased LV preload and filling.⁵¹ Aortic dimensions increased significantly during pharmacological challenge in our study, as described in human patients with hypertension, probably because of increases in BP.^57-60 Left ventricular PEP increased, which might be related to alterations in heart rate, conduction velocity, left ventricular end-diastolic pressure, left ventricular end-diastolic volume, afterload, and the inotropic state of the myocardium. The latter has been reported to be the dominant factor.⁶¹ Left ventricular PEP is influenced by the sympathetic activity of beta-1 adrenoreceptors and should therefore shorten under stimulation.⁶² The unexpected prolongation of LVPEP might be explained by increases in BP, which leads to a prolonged time to overcome the aortic pressure.^{63, 64} Increases in pulmonary dimensions after pharmacological challenge might also be caused by an increase in mean pulmonary arterial pressure, although this was not measured directly. These findings should be considered when examining horses with substantially increased heart rates or BP.

Repeatability of echocardiographic measurements in horses has been evaluated in a number of studies.^{1, 5, 6} Overall, 2D and M-mode measurements appeared to be more repeatable compared with other echocardiographic techniques.^{1, 5, 65} A clinically significant difference in echocardiographic variables should be larger than the 95% intermeasurement difference according to British Standards Institution (RC).⁴⁷ Statistically significant differences were identified for a number of variables after induction of physiological or pharmacological stress in our study, but many of these results do not represent clinically relevant differences because of the inherent measurement variabilities of the variables. However, mild increases in heart rate from 30 - 35 to 40 - 45 bpm may limit the effects of physiological stress in our study as heart rates might raise substantially during an echocardiographic exam under physiological stress. In addition, our study sample included healthy warmblood horses without an audible cardiac murmur or substantial valvular regurgitation. In cases with valvular regurgitation, small differences in HR and BP might contribute to significant changes in echocardiographic variables as the degree of valvular regurgitation is heavily influenced by variations in heart rate and BP.^{66, 67} Further research is needed to investigate the influence of small changes in HR and BP on echocardiographic dimensions in horses with valvular disease.²⁵ Echocardiographic assessment of horses under hospital conditions might result in higher HR and BP alterations which mimic those obtained in our study during the pharmacological challenge. This could result in clinically significant differences in echocardiographic measurements.

Our study has several limitations. An a priori power analysis was conducted. The sample size was sufficient to demonstrate a 5% to 10% difference in cardiac chamber dimensions, depending on the variability of the different measurements. Therefore, differences of <5% could have been missed. However, these differences were deemed to be clinically insignificant. Increases in HR caused by physiological stress could in some horses only be achieved by forward or backward walking of the horse during the echocardiographic examination. This might have influenced image quality compared with the examinations at rest or during pharmacological challenge. In our experiment, BP was measured noninvasively and correlations with invasive measurement techniques have not been performed. A percentage of error (15%-28%) has been reported between noninvasively and invasively BP measurements for standing horses.⁶⁸ However, aim of our study was to estimate and differentiate trends in BP among scans and not to assess BP most precisely or accurately. For this purpose, the used technique and type of device have been validated in horses.⁶⁹ The order in which image acquisition was performed during the pharmacological challenge could have had an influence on the results as images acquired shortly after drug administration may be more prone to drug effects than those recorded later on. Therefore, image acquisition was performed in random order. Furthermore, the physiological stress scan was performed at only mildly increased HR. In a clinical setting, HR might rise substantially during an echocardiographic exam and exceed 45 bpm, which could lead to significant differences compared with rest. This situation could be more comparable to the pharmacological challenge, although BP might not rise as substantially. Hydration state of the horses was not assessed during this crossover study, although hydration deficits or hypovolemia could have affected echocardiographic variables during the different scans.⁷⁰ However, none of the horses showed clinical signs of dehydration during clinical examinations before the scans and hydration deficits were assumed to be unlikely. A few reported echocardiographic variables are prone to error or limitations. Assessment of RV echocardiographic variables is influenced by respiration as well as apical trabeculation. Therefore, RV measurements appear generally to be less reliable. Decloedt et al. (2015) demonstrated a low variability for most 2D and M-mode echocardiographic variables (CV < 15% for RVIDd_{M_ch} and RVIDs_{M_ch}).²⁰ Our study showed low to moderate CV's for RV M-mode variables. Volumetric variables are calculated variables based on geometric assumptions and subsequently prone to error.⁵¹ PW Doppler–derived LV SV and CO are calculated estimates based on aortic dimensions and outflow velocity profiles, which are highly angle dependent. Inadequate alignment, variation in cursor placement, and inaccuracies in the determination of the flow area predispose to error. PW Doppler echocardiographic variables are therefore generally regarded to be less reliable and accurate.³⁹

In conclusion, physiological stress with mildly increased HR (40-45 bpm) resulted in increases in ventricular peak radial wall motion velocity during atrial contraction, reflecting an increased left atrial pump function. Most other left atrial, left ventricular, aortic and pulmonary 2D, M-mode, or Ao PWD measurements did not differ significantly with mildly increased HR (40-45 bpm) in healthy horses. This finding emphasizes an important influence of HR on measures of atrial contractile function. Pharmacological challenge with NBBM led to decreased left atrial and diastolic ventricular dimensions and increased aortic and pulmonary dimensions.

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

Study (EC 2020/100) was performed after approval and following the guidelines of the Ethical Committee of the Faculty of Veterinary Medicine, Ghent University, Belgium.

HUMAN ETHICS APPROVAL DECLARATION

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