Heart rate variability and quality of life in dogs with mitral valve disease treated with metoprolol

Copyright Beluque et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution and reproduction in any medium provided the original work is properly cited. Heart rate variability and quality of life in dogs with mitral valve disease treated with metoprolol Variabilidade da frequência cardíaca e qualidade de vida em cães com doença valvar mitral tratados com metoprolol


Introduction
Mitral valve disease (MVD) is the most prevalent cardiovascular disease in dogs, affecting up to 42.6% of animals over seven years of age (Santos et al., 2013). This progressive condition often leads to congestive heart failure (CHF), resulting in low cardiac output and hypotension (Triposkiadis et al., 2009).
One of the first responses to reductions in cardiac output and blood pressure is altered autonomic function, which consists of hyperactivity of the sympathetic autonomic nervous system (SANS), to the detriment of the parasympathetic autonomic nervous system (PANS) (Triposkiadis et al., 2009). When cardiac output is impaired, relative hypotension triggers an increase in serum norepinephrine concentration (Jensen et al., 2014), which acts on α1-adrenergic receptors to augment systemic vascular resistance as well as on β1-adrenergic receptors, exercising positive inotropic and chronotropic activity (Cunningham, 2004).
Activating the SANS initially benefits the cardiovascular system, but as sympathetic activity becomes chronic several deleterious effects are observed, such as tissue hypoperfusion, left ventricular dysfunction, and greater reduction of cardiac output (Frigerio & Roubina, 2005;Mehta et al., 2003).
In humans, treatment of CHF with beta-blockers has been proven beneficial in combating the harmful effects of chronic SANS activation. The benefits of treatment are decreased global left ventricular remodeling, reduced risk of hospitalization, shorter hospital stays (Klapholz, 2009), improved quality of life, and increased survival (Frigerio & Roubina, 2005;Gheorghiade et al., 2003). In contrast, research on the benefits of beta-blockers to treat dogs with endocardiosis is relatively scarce and does not permit definitive conclusions about the effectiveness of this therapy.
The degree of heart rate modulation by the autonomic nervous system (ANS) is most effectively assessed by heart rate variability (HRV). This, in turn, is obtained from long-term electrocardiography (Holter monitoring) (Vanderlei et al., 2009), which is also useful for diagnosing arrhythmias (Rasmussen et al., 2011). More advanced cases of CHF are less influenced by the PANS, with greater interference from the SANS. For this reason, HRV analysis is a very important and accessible tool for understanding the influence of the ANS in the various stages of CHF (Vanderlei et al., 2009).
Due to the consequences of SANS activation, this study was designed to evaluate the effect of metoprolol, a selective β1 beta-blocker, on HRV and quality of life in dogs with severe MVD.

Samples
Between December 2017 and November 2018, we selected dogs with stage C MVD according to the American College of Veterinary Internal Medicine (ACVIM) consensus guidelines (Atkins et al., 2009). After completion of the study, a new consensus for the diagnosis and treatment of MVD was published (Keene et al., 2019). The patients selected for our study also met the criteria for classification as stage C of the last consensus, since all were symptomatic.
All animals were attended at the Veterinary Hospital of the Faculdade de Ciências Agrárias e Veterinárias (FCAV) -Universidade Estadual Paulista (UNESP), the owners were informed about the study, and after agreeing to include the animals in the experiment signed terms of consent. The study was approved by institutional ethics committee for animal research at FCAV-UNESP three months before research began (protocol number 012225/17).
The echocardiographic inclusion criteria were left atrium/aorta ratio (LA/Ao) ≥1.6 and left ventricular internal diameter in diastole (LVIDd) above the reference values according to the weight of the animals (Boon, 2011). In the electrocardiographic assessment, animals with sinus arrhythmia, sinus rhythm, or sinus tachycardia without conduction blocks were included. All qualidade de vida após instituição do metoprolol. Considerando os resultados obtidos é possível inferir que o tratamento com o betabloqueador em cães com doença valvar mitral pode ser considerado uma ferramenta terapêutica adicional na terapia da insuficiência cardíaca congestiva classe C.
Patients were followed for at least two months. Exclusion factors were concomitant illnesses such as endocrine disorders, liver disease, kidney disease, neoplasms, and infectious diseases. Dogs were evaluated at the beginning of treatment (T0) and again 30 days after the start of CHF therapy and the start of therapy with metoprolol (T1) and 30 days after the start of therapy with the beta-blocker (T2).

Systolic blood pressure assessment
All patients underwent non-invasive systolic blood pressure measurement with the Doppler method (Doppler Flow Detector 812, Oregon, USA) at T0, T1 and T2. Five measurements were taken; the first two were excluded, and the average of the three remaining measurements was calculated. Cuff size was selected to represent 30-40% of the circumference of the patient's thoracic limb (Acierno et al., 2018); either right or left was permissible, but the same limb was used for all measurements (T0, T1, and T3). Patients were kept in the same decubitus position during all measurements. For more restless patients, blood pressure was measured as they sat on the examination table or with the guardian, maintaining the same position during each of the three measuring times (T0, T1, and T3).

Echocardiographic assessment
All patients were evaluated at T0, T1, and T2 using the same echocardiographic device (MyLabTM30 Gold Cardiovascular, Esaote, Italy). Patients were maintained in lateral decubitus to obtain cross-sectional and longitudinal images of the right parasternal window and apical images of the left parasternal window (Boon, 2011). Mitral peak early diastolic filling velocity (E wave) was obtained, as well as early diastolic mitral annular velocity (E') derived from tissue Doppler imaging (TDI), left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs), ejection fraction (EF), fractional shortening (FS) and isovolumetric relaxation time (IVRT), and qualitative assessment of all chambers, myocardium and pericardium.

Long-term electrocardiogram (Holter monitor)
Adhesive electrodes were positioned on the shaved skin surface to capture signals. Two negative electrodes were placed, one on the right chest wall and the other in the manubrium region, and two positive electrodes were also positioned, one on the left chest wall and the other on the xiphoid cartilage. A bandage was applied around the thorax, and the dogs were fitted with a vest to hold the three-channel Cardiolight digital recording device (Cardios Sistemas, São Paulo, Brazil). The animals subsequently wore the device for 24 hours in the owner's home, a familiar environment. This procedure was followed for assessment at T1 and T2; it was not possible to perform this exam at T0 because owners were unwilling to wait for another exam or return over the following days with the Holter monitor.
The cardiac cycles were analyzed using Cardio Manager S540 software (Cardios Sistemas). Rates of HRV assessed over time included standard deviation of all RR intervals (SDNN), standard deviation of the mean RR intervals obtained every five minutes (SDANN), mean standard deviation of the RR intervals obtained every five minutes (SDNNIDX), mean square root of the square of the differences between adjacent RR intervals (rMSSD), percentage of consecutive R-R intervals lasting more than 50 ms (pNN50), and mean duration of RR intervals (mean NN). Normal QRS complexes and arrhythmic events (supraventricular and ventricular) were initially determined by the software, and all morphological patterns and arrhythmias were subsequently corrected by the same person who performed the exams.

Quality of life assessment
The Functional Evaluation of Cardiac Health (FETCH questionnaire) (Freeman et al., 2005) was used to assess quality of life. Owners answered the 17 questions on the questionnaire before the dogs started treatment with metoprolol (T1) and again after one month of treatment with metoprolol (T2). The responses to the questionnaire use a scale of 0 to 5, with 0 indicating absence of the clinical sign and 5 indicating maximum expression.

Therapeutic protocol
After ruling out concomitant diseases, therapy for CHF was initiated with enalapril maleate (0.5 mg/kg once daily), furosemide (1-2 mg/kg twice daily), and spironolactone (2 mg/kg once daily) at T0. For patients already undergoing treatment with these medications, appropriate dose adjustments were made. The majority (5/8) of the patients were already receiving treatment for CHF, one was treated with only enalapril (1/8), and others received enalapril and furosemide (3/8). The only dog followed at the Veterinary Hospital (FCAV-UNESP) was treated with enalapril, furosemide, spironolactone, and pimobendan for over six months, and the remaining animals were not being treated (3/8). The dog treated with enalapril had received medication for over a year, while the dogs treated with enalapril and furosemide had been receiving medication for a period of 2 (2/8) and 8 (1/8) weeks, respectively. Dogs with systolic dysfunction (5/8) documented by echocardiography also received pimobendan (0.25-0.35 mg/kg twice daily). Patients were reevaluated 30 days after the start of therapy, and metoprolol (0.2 mg/kg twice daily) was then added to the regime.

Statistical analysis
The data obtained were subjected to the Shapiro-Wilk normality test. When Gaussian distribution was found, means and standard deviations were calculated and the paired Student's T test or ANOVA was used for repeated measures, followed by the Tukey multiple comparisons test for comparison between different evaluation times (T1 and T2). The chi-square test was used to analyze the questionnaire. All statistical analyses were performed with R software, and P <0.05 was considered significant.
All evaluations (for example, acquisition of echocardiographic imaging, applying questionnaire to owners, electrocardiography, measurement of systolic blood pressure, and interpreting exams) were performed by the same person (T.B.).

Results
Initially, 18 patients were selected over a period of ten months, but six were lost to follow-up because of the distance to the hospital. Two patients had concomitant illnesses during the experiment, one patient died at the owner's home before starting treatment with metoprolol (unknown cause), and one patient did not continue because the owner was unable to administer the medications properly.
Most of the patients were alive for months after the end of the study. Only one patient had acute pulmonary edema 3 months after the end of the study and died at the owner's home a few hours later.
A significant increase was observed in FS (P=0.02) at T2 when compared to T0, as well as in EF (P=0.0008) ( Table 1). There was also a significant decrease in LVIDs (P=0.002) and E wave velocity (P=0.02) in T2 when compared to T0 (Table 1). There was a statistically significant difference (P=0.006) in the LA/Ao ratio in T1 and T2, but there was no statistical difference in the values of LA (P = 0.41) and Ao (P=0.43) when evaluated individually (Table 1).
The HRV parameter data for dogs before and after treatment with beta-blockers are shown in Table 2. There was a statistically significant increase for pNN50 between T1 and T2 (P=0.039). The mean value for NN increased at T2 (Table 2), but the difference was not statistically significant (P=0.056). No significant differences were seen in the mean values for minimum HR (P=0.52), mean HR (P=0.06), or maximum HR (P=0.55) at T2. Supraventricular and ventricular arrhythmias before and after treatment are presented in Table 3. Four patients (50%) had ventricular premature complexes at the first long-term electrocardiogram and four (50%) had supraventricular premature complexes, while only three (37.5%) exhibited both concomitantly. All arrhythmias presented were isolated; no statistically significant change was seen in supraventricular or ventricular arrhythmias at T2 (P=0.296 and P=0.946, respectively).  There was a statistically significant reduction in the FETCH questionnaire score for clinical signs in 37.5% (3/8) of patients when questionnaires were compared individually between T1 and T2 (P=0.0004; P=0.007; P=0.006). Approximately 66% (2/3) of the dogs with reductions in FETCH scores also presented an increase in pNN50 values in T2; however, because there was not sufficient data for testing, it was not possible to assess statistical significance.

Discussion
Prior to therapy with metoprolol, the mean value for pNN50 (shown in Table 2) was similar to the mean value obtained in another study (Oliveira et al., 2012) in dogs with stage C and D MVD. This value was also significantly lower than the mean for healthy patients and patients with MVD without CHF evaluated in this same study, showing that patients with CHF are less influenced by the PANS (Oliveira et al., 2012), which was also corroborated by an additional study (Spier & Meurs, 2004). The mean value for pNN50 increased after therapy with metoprolol and reached the values for healthy animals in other studies (Bogucki & Noszczyk-Nowak, 2015;Oliveira et al., 2012).
HRV parameters in the time domain express the difference between consecutive RR intervals; for example, pNN50 reflects the predominance of vagal action (Bogucki & Noszczyk-Nowak, 2017;Sztajzel, 2004;Zacché et al., 2017), making pNN50 an excellent indicator for assessing the influence of PANS, as in this study. A study in human patients evaluated PANS activity according to pNN50 and found higher values after vagus nerve stimulation (Liu et al., 2017). Consequently, the increase in pNN50 after metoprolol therapy indicates higher HRV, in other words, greater participation of the PANS in the autonomic balance, similar to what occurs in healthy patients (Bogucki & Noszczyk-Nowak, 2015;Oliveira et al., 2012). Although not significant, an increase was seen in the mean NN value after treatment with the beta-blocker, a parameter that also reflects vagal action. It is possible that the short period of treatment with the beta-blocker was not sufficient to permit greater changes in the mean NN between T1 and T2; longer courses of therapy could show more significant changes.
Studies have indicated pNN50 is a good parameter for assessing autonomic balance (Bogucki & Noszczyk-Nowak, 2017;Oliveira et al., 2012), which demonstrates the importance of the statistically significant difference in this variable observed in this present study. Other authors have used this variable to demonstrate more intense performance of the PANS in healthy dogs at rest compared to those that performed physical activity (Blake et al., 2018), which leads us to question whether the effects of metoprolol would have been different in patients with moderate physical activity compared to patients with mild physical activity. Since this was beyond the scope of this present study, activity intensity and duration of the dogs was not evaluated, but could be assessed in future studies. PANS activity was also higher in dogs during various sleep phases in relation to waking periods (Bálint et al., 2019), confirming the need to evaluate electrocardiography over the 24-hour period, because the SANS could have prevailed if the evaluation was performed only when the dogs were most active. Conversely, the PANS could been more dominant if the Holter monitor was in place only when the dogs tended to sleep, altering the real values.
Furthermore, in a comparison between calm dogs and dogs with a history of aggression, more aggressive dogs had lower pNN50 values (Craig et al., 2017); this result is interesting, since two owners reported their dogs had been more irritable for a few months, and that irritability decreased considerably after treatment with metoprolol. For one of these dogs, quality of life according to Table 3. Arrhythmias observed during long-term electrocardiography in dogs (n=8) with stage C mitral valve disease before (T1) and 30 days after (T2) initiation of therapy with metoprolol. the questionnaire improved significantly. Further studies are needed to further investigate the relationship between aggressiveness or irritability and metoprolol in dogs.

Parameter
Higher pNN50 values were also observed in healthy patients compared to dogs with MVD (Pirintr et al., 2017), in Cavalier King Charles Spaniels without CHF compared to patients already in CHF (Rasmussen et al., 2011), in dogs with advanced MVD without a history of syncope compared to dogs with a history of syncope, and in healthy adolescents compared to those with mitral valve prolapse (Olexova et al., 2020). All these studies cited indicated the lower participation of PANS in patients with MVD, and even lower in cases of CHF. Such findings indicate the importance of studies assessing HRV in dogs with MVD, especially symptomatic animals. Future studies may more concisely define the prognostic role of Holter variables for MVD, and compare them with patients treated with beta-blockers.
In the present study, metoprolol therapy yielded similar values for HRV in the treatment group compared to those of healthy patients included in other studies (Bogucki & Noszczyk-Nowak, 2015;Oliveira et al., 2012), an outcome that was not observed when these patients were only treated with angiotensin-converting enzyme inhibitors, diuretics, and inodilators. Another study (Chompoosan et al., 2014) evaluated pNN50 in patients with MVD before and after 14 days of therapy with enalapril maleate, and found no significant change in this parameter. This finding could corroborate our statement that metoprolol made a major contribution to the increase in HRV, since in the study cited there was no interference with the use of the vasodilator alone. However, the effect of therapy with enalapril associated with a diuretic and inodilator on HRV for a longer period (as in our study) has not been evaluated, and may also have influenced the results observed.
A slight reduction in mean HR was seen after administration of the beta-blocker, although it was not significant. In stabilization of heart disease, the SANS is known to become less stimulated (Rasmussen et al., 2012), and as a result reduced HR is expected.
Interestingly, we found no reduction in arrhythmias one month after starting beta-blocker therapy. The increase in ventricular arrhythmias, although not significant, is believed to result from the evolution of the disease. Valve prolapse can mechanically stimulate the endocardium, and traction exerted by the leaflets on the chordae tendinae and the stretching of the ventricular wall are recognized mechanisms that trigger ventricular arrhythmias (Crosara et al., 2010). This would also explain the greater frequency of ventricular arrhythmias compared to supraventricular arrhythmias. However, the data from this present study contrast with findings from a more recent study (Oliveira et al., 2014) that demonstrated a higher prevalence of supraventricular arrhythmias in dogs with MVD compared to ventricular arrhythmias, and the group of dogs with CHF had a statistically higher number of supraventricular arrhythmias than the group without CHF and the group of healthy dogs.
The quality of life assessment for the patients treated with metoprolol showed significant improvement in more than a third after they started beta-blocker therapy. Two-thirds of the dogs with improved quality of life (according to the FETCH criteria) exhibited changes in HR variability indicative of greater participation by the PANS, represented by an increase in pNN50, although this was not statistically significant. Previous studies in humans have also shown improved quality of life in patients treated with metoprolol (Frigerio & Roubina, 2005;Gheorghiade et al., 2003).
All animals remained stable throughout the study period, and as of their last visit to the veterinary hospital no deterioration was seen in echocardiographic, electrocardiographic, clinical, or SBP parameters. Only one animal died while receiving treatment, 3 months after the end of the study. The dog was a patient of the cardiology service, maintained the routine of return visits, and attended the last appointment a week before death. No systolic dysfunction, hypotension, or any other alteration that could justify this death with the use of the prescribed treatment was seen up to the time of the last visit, nor were there changes indicating CHF progression, since none of the echocardiographic parameters were compatible with the indices of congestion described (Morgan et al., 2020;Schober et al., 2010), and no pattern of congestion was seen in the radiographic evaluation (Oui et al., 2015). A necropsy was not performed since the animal lived in a different city than the study site, but the history of acute dyspnea described by the guardian and the clinical reassessment and recent examinations without worsening of CHF suggest a rupture of the chordae tendineae.
A main concern surrounding beta-blockers is a potential negative effect on hemodynamic balance, since the reduction in HR and inotropism caused by the blockade in β receptors, along with potential blockade in alpha receptors (Plumb, 2011), could further decrease cardiac output (CO). However, the SBP of the dogs in this study remained stable throughout the evaluation period, and no clinical signs compatible with reduced cardiac output were reported by the tutors, nor were there changes in physical examination. This may be due to the selectivity of metoprolol by β receptors (Plumb, 2011), and the safe dose that was used.
Although methods with greater sensitivity for assessing left ventricular systolic function were not utilized, the increase in FS and EF associated with the decrease in LVIDs show that systolic function was not impaired. This was similar to the results obtained in another study (Zacà et al., 2009) that also evaluated the effects of metoprolol in dogs and obtained statistically higher values for EF and lower values for LVID. The minimum dose of metoprolol used may have contributed to these findings in our study, even in patients previously diagnosed with systolic dysfunction as well as patients who did not receive pimobendan.
The present study had some limitations which are common in studies involving clinical patients and owners, but should be considered when interpreting the results. The sample size was small, since over 50% of the patients initially recruited were removed from the study and stage C is not the predominant stage of MVD in our cardiology department. Furthermore, the more advanced age at which patients with MVD usually present clinical signs is also conducive to the onset of other diseases, which excluded many patients. The study period was short, since most owners lived in other cities and were unable to return more often. All patients were selected at the same veterinary hospital, which allows the possibility of bias. Finally, there was no concomitant treatment with a placebo to replace metoprolol, and only one stage of valve disease was used. More evident benefits may be found in other stages of MVD, especially for more extensive therapies and in studies involving larger numbers of patients.
The use of metoprolol should also be evaluated in other classes of MVD in double-blind, placebo-controlled studies while other medications that keep the patient's clinical condition stable are maintained. Assessments involving larger samples and longer treatment periods could demonstrate favorable outcomes from the use of beta-blockers in CHF therapy, especially in patients demonstrating autonomic imbalance during long-term electrocardiography. A longer study period could provide important information on patient survival with and without administration of beta-blockers.

Conclusions
Our study demonstrated that 30 days of metoprolol associated with at least 2 months of classic treatment for severe mitral valve disease improved quality of life and cardiac autonomic function in dogs.

Ethics statement
The responsible for the non-human animals formally consented to the study. The study was approved by ethics committee for the use of animals of Faculdade de Ciências Agrárias e Veterinárias (FCAV) of Universidade Estadual Paulista (UNESP), with protocol number 012225/17.