Prognostic value of pulmonary vascular resistance estimated by echocardiography in dogs with myxomatous mitral valve disease and pulmonary hypertension

Abstract

Background

Progression to combined post‐ and pre‐capillary pulmonary hypertension (PH) provides prognostic information in human patients with post‐capillary PH. Pulmonary vascular resistance estimated by echocardiography (PVRecho) is useful for the stratification of dogs with myxomatous mitral valve disease (MMVD) and detectable tricuspid regurgitation.

Objectives

To evaluate the prognostic value of PVRecho in dogs with MMVD.

Animals

Fifty‐four dogs with MMVD and detectable tricuspid regurgitation.

Methods

Prospective cohort study. All dogs underwent echocardiography. The PVRecho was calculated based on tricuspid regurgitation and the velocity‐time integral of the pulmonary artery flow. To evaluate the influence of echocardiographic variables on cardiac‐related deaths, Cox proportional hazard analysis was performed. Additionally, Kaplan‐Meier curves classified by PVRecho tertiles were made and compared using log‐rank tests to evaluate the influence of PVRecho on all‐cause mortality and cardiac‐related death.

Results

The median follow‐up time was 579 days. Forty‐one dogs with MMVD (PH severity [number]: no or mild, 21/33; moderate, 11/11; severe, 9/10) died during the study. In the multivariable Cox proportional hazard analysis adjusted for age, sildenafil administration, and American College of Veterinary Internal Medicine stage of MMVD, left atrial to aortic diameter ratio and PVRecho remained significant (adjusted hazard ratio [95% confidence interval]: 1.2 [1.1‐1.3] and 2.1 [1.6‐3.0], respectively). Higher PVRecho showed a significant association with lower survival rates.

Conclusions and Clinical Importance

Left atrial enlargement and high PVRecho were independent prognostic factors in dogs with MMVD and detectable tricuspid regurgitation.

Abbreviations

3seg

only right ventricular free wall analysis

6seg

global right ventricular analysis

AUC

area under the curve

CI

confidence interval

E

early‐diastolic transmitral flow velocity

E/A

early‐diastolic to late‐diastolic transmitral flow velocity ratio

e'

tissue Doppler imaging‐derived peak early‐diastolic myocardial velocity of the septal mitral annulus

HR

hazard ratio

LA/Ao

left atrial to aortic diameter ratio

LVIDDN

end‐diastolic left ventricular internal diameter normalized by body weight

LVIDSN

end‐systolic left ventricular internal diameter normalized by body weight

MMVD

myxomatous mitral valve disease

PH

pulmonary hypertension

RHF

right heart failure

ROC

receiver operating characteristic

RV FACn

RV fractional area change normalized by body weight

RV s'

tissue Doppler imaging‐derived peak systolic myocardial velocity of the lateral tricuspid annulus

RV

right ventricular

RVEDA index

end‐diastolic RV area normalized by body weight

RVESA index

end‐systolic RV area normalized by body weight

RVIDd index

end‐diastolic RV internal dimension normalized by body weight

RV‐SL

RV longitudinal strain

RV‐SrL

RV longitudinal strain rate

TAPSEn

tricuspid annular plane systolic excursion normalized by body weight

TR

tricuspid regurgitation

1. INTRODUCTION

Pulmonary hypertension (PH) is a common complication of myxomatous mitral valve disease (MMVD), the most common cardiac disease in dogs. ¹ , ² , ³ Pulmonary hypertension caused by left‐heart disease is called post‐capillary PH and is hemodynamically classified into 2 subtypes: isolated post‐capillary PH and combined post‐ and pre‐capillary PH (Cpc‐PH). ¹ , ² Isolated post‐capillary PH is caused solely by elevated left‐atrial pressure, whereas concurrent pulmonary vascular remodeling increases pulmonary vascular resistance (PVR) and leads to progression to Cpc‐PH. Therefore, the pulmonary arterial pressure of Cpc‐PH is greater than that of isolated post‐capillary PH. As the progression to Cpc‐PH is associated with a poor prognosis in human patients with post‐capillary PH, ⁴ , ⁵ , ⁶ , ⁷ , ⁸ detection of increased PVR is essential for evaluating the pathophysiology and prognoses of PH in dogs with MMVD.

Right‐heart catheterization is currently the best method for assessing right‐heart hemodynamics, including pulmonary arterial pressure and PVR. ¹ , ⁹ However, performing this technique in all dogs is impractical because of the need for special equipment and the high anesthetic risk for dogs with progressive heart disease. Therefore, noninvasive alternatives are necessary in clinical veterinary practice. In humans, the PVR estimated using echocardiography (PVRecho), which is calculated using the tricuspid regurgitation (TR) velocity and the velocity‐time integral of the pulmonary artery flow (PV VTI), can detect increased PVR without using right‐heart catheterization. ¹⁰ , ¹¹ , ¹² , ¹³ The PVRecho has utility in the diagnosis and stratification of PH in dogs with MMVD. ¹⁴

Several studies have investigated the prognostic value of various echocardiographic variables in dogs with MMVD and post‐capillary PH. ³ , ¹⁵ , ¹⁶ However, no study has evaluated the prognostic value of PVRecho in dogs. This study evaluated the prognostic role of PVRecho in dogs with MMVD at various stages (American College of Veterinary Internal Medicine [ACVIM] stage B1, B2, and C/D) and detectable TR. We hypothesized that PVRecho could provide additional prognostic information for dogs with MMVD and Cpc‐PH.

2. MATERIALS AND METHODS

This was a prospective cohort study. All procedures followed the Guidelines for Institutional Laboratory Animal Care and Use, and the study was approved by the Ethical Committee for Animal Use (approval number: R2‐5). Written informed consent that authorized the participation of each dog in the study was obtained from their owner.

2.1. Animals

The study prospectively included client‐owned dogs with MMVD and detectable TR regardless of the presence of left heart remodeling and left heart failure that presented to our university medical teaching hospital in Japan from October 2017 to May 2019. All dogs underwent complete physical examinations, electrocardiography, oscillometric method‐derived blood pressure measurements, transthoracic radiography, and echocardiographic examinations. The MMVD was clinically diagnosed using transthoracic echocardiography based on the presence of mitral valve thickening, prolapse, or a combination of these, and associated mitral regurgitation. ¹⁴ , ¹⁷ , ¹⁸ , ¹⁹ The TR was evaluated by inspecting the tricuspid valve from multiple viewing angles and using color Doppler echocardiography; the peak TR velocity was obtained using the continuous‐wave spectral Doppler method. Dogs were excluded from the study if they had (a) diseases other than MMVD that might increase pulmonary arterial pressure (eg, respiratory disease, thromboembolic disease, heartworm disease, neoplastic disease, and endocrine disease), ¹ (b) systemic hypertension (systolic systemic blood pressure >160 mm Hg), ²⁰ (c) sustained arrhythmia which might influence hemodynamics (eg, atrial fibrillation, ventricular tachycardia, and sinus arrest), and/or (d) missing data. To differentiate dogs with diseases that met the excluding criteria, dogs additionally underwent radiography in the inspiratory and expiratory phases, upper airway radiography, fluoroscopic examination, blood chemistry examinations, abdominal ultrasonography, and blood coagulation tests. Dogs with abnormally high concentrations of fibrin degradation products (>4.0 μg/mL) and D‐dimers (>2.0 μg/mL) were excluded from the study because of suspected pulmonary artery thromboembolism.

Dogs with MMVD were divided into 1 of 3 groups, based on the ACVIM consensus. Asymptomatic dogs with no or minimal remodeling were classified as stage B1; asymptomatic dogs with significant left‐heart remodeling based on an end‐diastolic left‐ventricular internal dimension normalized using body weight (LVIDDN) >1.7 and left‐atrium to aortic diameter ratio (LA/Ao) >1.6, were classified as stage B2; and symptomatic dogs with current or past clinical signs of heart failure caused by MMVD were classified as stage C/D. ²¹ Right heart failure (RHF) was diagnosed based on the presence of at least 1 radiographic or ultrasonographic finding indicative of ascites, pleural effusion, or pericardial effusion without any abnormality other than PH that may be responsible. ¹⁴ , ¹⁸ , ¹⁹ Furthermore, dogs were classified into 1 of the 3 PH severity groups based on the TR velocity reported in the previous studies. ¹⁸ , ²² , ²³ Dogs that had TR <3.5 m/s were classified as no or mild PH; dogs that had TR 3.5‐4.3 m/s were classified as moderate PH; dogs that had TR >4.3 m/s were classified as severe PH. ¹⁸ , ²² , ²³

2.2. Echocardiography

Conventional, 2‐dimensional, and Doppler echocardiography were performed by a single investigator using the Vivid E95 echocardiographic system (GE Healthcare; Tokyo, Japan). Lead II ECG was simultaneously recorded and displayed. All data were recorded for at least 5 consecutive cardiac cycles from non‐sedated dogs that were restrained manually in right and left lateral recumbency. All echocardiographic measurements were performed using an EchoPAC version 204 workstation (GE Healthcare; Tokyo, Japan) by another observer who had been trained by a cardiologist and was blinded to the dogs' identities. The mean values obtained from high‐quality images of 3 consecutive cardiac cycles of the sinus rhythm were used for the statistical analyses.

To evaluate left‐heart morphology and function, the LA/Ao ratio, LVIDDN, end‐systolic left‐ventricular internal dimension normalized using body weight (LVIDSN), and fractional shortening were measured using the B‐mode method as previously described. ²⁴ , ²⁵ , ²⁶ In addition, Doppler echocardiography was performed to measure the early diastolic (E) and late diastolic (A) transmitral flow velocities; the tissue Doppler imaging‐derived peak early‐diastolic myocardial velocity of the septal mitral annulus (e') was also measured. ²⁷ The E, A, and e′ were obtained using the left apical 4‐chamber view. The E to A (E/A) and E to e′ (E/e′) ratios were then calculated.

For the indicators of RV morphology, end‐diastolic RV area (RVEDA), end‐systolic RV area (RVESA), and end‐diastolic RV internal dimension (RVIDd) were measured using the left apical 4‐chamber view optimized for the right heart (RV focus view), as described previously. ¹⁴ , ¹⁸ , ¹⁹ , ²⁸ , ²⁹ , ³⁰ , ³¹ The RVIDd was defined as the largest diameter at the middle right ventricle parallel to the tricuspid annulus. ¹⁸ , ³⁰ , ³¹ These variables were indexed by body weight using the following formulas: RVEDA index = (RVEDA [cm²])/(body weight [kg])^0.624, RVESA index = (RVESA [cm²])/(body weight [kg])^0.628, and RVIDd index = (RVIDd [mm])/(body weight [kg])^0.327. ³⁰ Tricuspid annular plane systolic excursion (TAPSE), RV fractional area change (FAC), and tissue Doppler imaging‐derived peak systolic myocardial velocity of the lateral tricuspid annulus (RV s') were measured as described previously. ¹⁴ , ¹⁸ , ¹⁹ , ²⁷ , ³⁰ , ³¹ , ³² The TAPSE was measured using the B‐mode method, as described previously. ¹⁸ , ³³ The normalized values for TAPSE and RV FAC (TAPSEn and RV FACn, respectively) were calculated using the following formulas: TAPSEn = (TAPSE [mm])/(body weight [kg])^0.284 and RV FACn = (RV FAC [%])/(body weight [kg])^−0.097. ²⁹ , ³³ The PVRecho was calculated using the TR velocity and PV VTI, as follows: PVRecho = (TR velocity [m/s])²/(PV VTI [cm]; Figure 1). ¹¹ , ¹² , ¹⁴ The TR was evaluated by inspecting the tricuspid valve from multiple views, and the TR velocity was defined as the highest velocity obtained from the dense TR flow profile. The PV VTI was measured by tracing the wave envelope of the Doppler spectrum of the pulmonary artery flow which was carefully obtained from the right parasternal short‐axis view at the level of the heart base optimized for pulmonary artery. ³⁴

FIGURE 1.

Representative data of pulmonary vascular resistance estimated by echocardiography of 54 dogs with myxomatous mitral valve disease and pulmonary hypertension. PV VTI, velocity‐time integral of the pulmonary artery flow; PVRecho, pulmonary vascular resistance estimated by echocardiography; TR, tricuspid regurgitation.

Two‐dimensional speckle‐tracking echocardiography was performed to measure the RV longitudinal strain (RV‐SL) and strain rate (RV‐SrL). These variables were obtained using the RV focus view and left‐ventricular 4‐chamber algorithms, as previously described. ¹⁴ , ¹⁸ , ¹⁹ , ³⁵ , ³⁶ RV free‐wall analysis (3seg) was performed by tracing the endomyocardial border from the lateral tricuspid annulus to the RV apex; global RV analysis (6seg) was performed by tracing the endomyocardial border from the lateral tricuspid annulus to the septal tricuspid annulus via the RV apex. When necessary, manual adjustments were performed to include and track the entire myocardium over the cardiac cycle. When the automated software could not track the myocardial regions, the regions of interest were retraced and recalculated. The RV‐SL was defined as the absolute value of the negative peak value of the global strain wave, which was calculated automatically. RV‐SrL was defined as the absolute value of the negative peak value of the global strain rate wave during systole. ¹⁴ , ¹⁸ , ¹⁹ , ³⁵ , ³⁶

The intra‐ and inter‐observer measurement variabilities of all the variables evaluated in this study have been reported in our previous report, in which all variables showed acceptable measurement variabilities. ¹⁴

2.3. Survival analysis

Survival information for all dogs was obtained by reviewing the medical records at our institution. When necessary, the study investigator contacted each dog's owner, primary care veterinarians, or both, to obtain survival information. If the dogs had died, the date and cause of death (cardiac‐related or non‐cardiac‐related) were recorded. Cardiac‐related death was defined as natural death/euthanasia because of congestive heart failure that was refractory to medical treatment. In this study, sudden death was also defined as cardiac‐related death if there was no other cause than cardiac diseases. Death with any other cause was defined as non‐cardiac‐related death. Survival time was calculated from the date of examination to the date of death for dogs that died within the study period (October 1, 2017 to September 30, 2021), or to the date of data collection for surviving dogs (September 30, 2021).

2.4. Statistical analysis

Statistical analyses were performed using EZR software version 1.41 (Saitama Medical Center, Jichi Medical University). ³⁷ Categorical data are presented as absolute numbers, and frequencies are presented as percentages. Continuous data are reported as medians with interquartile ranges.

The normality of the data was evaluated using the Shapiro‐Wilk test. Categorical data were compared according to PH severity using Fisher's exact test. Continuous data were compared among PH severity groups using 1‐way analysis of variance with subsequent pairwise comparisons using Tukey's multiple comparison test (for normally distributed data) or the Kruskal‐Wallis test with subsequent pairwise comparisons using the Steel‐Dwass test (for nonnormally distributed data). The prognostic values of echocardiographic variables were determined using univariable and multivariable Cox proportional hazard analyses. Variables with P < .10 in the univariable analyses were included in the multivariable analyses. The multivariable model was adjusted for age, sildenafil administration, and ACVIM stage. Spearman rank correlation analysis was performed to adjust for the multicollinearity of echocardiographic variables. The results of Cox proportional analyses are presented using the hazard ratio (HR) and the respective 95% confidence interval (CI). Assumptions of proportional hazards and model fit were verified using Schoenfeld residuals. To evaluate the influence of PVRecho on all‐cause mortality and cardiac‐related death, Kaplan‐Meier curves which were stratified according to the tertile of PVRecho and adjusted for age and sildenafil administration were constructed and compared using log‐rank tests with Bonferroni correction. Dogs that were still alive at the end of the study period (September 30, 2021) were right‐censored; dogs that were lost during follow‐up were also right‐censored on the day of final contact. Statistical significance was set at P < .05 for all analyses.

3. RESULTS

3.1. Clinical profiles

Seventy‐two dogs with MMVD and detectable TR met the inclusion criteria. Eighteen dogs were excluded (Figure 2) for a total of 54 dogs enrolled in the study, including of 33 dogs with no or mild PH, 11 dogs with moderate PH, and 10 dogs with severe PH. The study population consisted of the following breeds: Chihuahua (n = 12, 22%), Toy Poodle (n = 6, 11%), mixed breed (n = 5, 11%), Shi Tzu (n = 4, 7%), Miniature Dachshund (n = 3, 6%), Maltese (n = 3, 6%), Miniature Schnauzer (n = 3, 6%), Papillon (n = 2, 4%), Pomeranian (n = 2, 4%), Cavalier King Charles Spaniel (n = 2, 4%), Chinese crested dog (n = 2, 4%), Norfolk Terrier (n = 2, 4%), and 1 dog each from 8 other breeds. Table 1 shows the clinical data and oral medications of our study cohorts at study inclusion classified by PH severity. Dogs with severe PH were older than those with no or mild PH. Body weight, sex, and systemic blood pressure showed no significant differences among PH severity groups. Regarding the ACVIM stage of MMVD, all dogs with Stage B1 had no or mild PH; 19/24 (79%) of Stage B2 dogs had no or mild PH, 4/24 (17%) had moderate PH, 1/24 (4%) had severe PH; 6/22 (27%) of Stage C/D dogs had no or mild PH, 7/22 (32%) had moderate PH, 9/22 (41%) had severe PH. No dog with Stage B1 had moderate to severe PH. In this study, 4/22 (18%) of Stage C/D dogs (2 dogs with moderate PH and 2 dogs with severe PH) showed radiographic evidence of active pulmonary edema at study inclusion. Eight dogs were diagnosed with RHF at the time of study inclusion: 4 dogs showed ascites (ascites of 2 dogs were proven to be transudate); 2 dogs showed pleural effusion (both effusions were proven to be transudate); 1 dog showed ascites and mild pericardial effusion; and 1 dog showed transudative pericardial effusion. Although 6/8 (75%) dogs with RHF had severe PH, 2/8 (25%) had moderate PH based on TR velocity (both TR velocities were 3.5 and 3.6 m/s). Seventy‐four percent of dogs were managed with various cardiac drugs, including angiotensin‐converting enzyme inhibitors, pimobendan, loop diuretics, sildenafil, or a combination of these. Significantly greater percentages of dogs with RHF and those receiving oral medications were observed in dogs with severe PH. Electrocardiography revealed that 8/54 dogs (15%; 1 dog with no or mild PH, 2 dogs with moderate PH, and 5 dogs with severe PH) had a single supraventricular extrasystole and 3/54 dogs (6%; 1 dog with no or mild PH and 2 dogs with severe PH) had a single ventricular extrasystole. However, no dogs were observed to have sustained arrhythmias which might influence hemodynamics. Clinical data and oral medications classified by PVRecho tertiles are presented in Table S1.

FIGURE 2.

Flow diagram representing the eligible dogs with myxomatous mitral valve disease (MMVD) and detectable tricuspid regurgitation (TR) and reason of exclusion.

TABLE 1.

Clinical data at study inclusion of 54 dogs with myxomatous mitral valve disease and detectable tricuspid regurgitation.

Variables	PH severity (n = 54)			P *
	No or mild (n = 33)	Moderate (n = 11)	Severe (n = 10)
Age (year)	11.2 (9.3‐13.3)	12.5 (12‐13.2)	14.3 (13.3‐14.7) a	.02
Body weight (kg)	4.6 (3.3‐6.9)	5.9 (3.3‐7.3)	3.3 (2.3‐7.2)	.46
Sex (male/female)	14/19	6/5	5/5	.74
ACVIM stage (B1/B2/CD)	8/19/6	0/4/7	0/1/9	<.01
RHF (yes/no)	0/33	2/9	6/4	<.01
Systolic blood pressure (mm Hg)	133 (114‐146)	136 (116‐149)	130 (104‐142)	.48
Mean blood pressure (mm Hg)	96 (83‐108)	98 (85‐111)	93 (80‐110)	.72
Cardiovascular drugs (yes/no)
Angiotensin converting enzyme inhibitor	22/11	9/2	9/1	.34
Pimobendan	10/23	8/3	9/1	<.01
Loop diuretics	2/31	1/10	3/7	.09
Sildenafil	0/33	1/10	6/4	<.01

Note: Categorical and continuous data are expressed as absolute numbers and medians (interquartile range), respectively.

Abbreviations: ACVIM, American College of Veterinary Internal Medicine; PH, pulmonary hypertension; RHF, right heart failure.

^*For categorical data, P‐values of Fisher's exact test. For continuous data, P‐values of 1‐way analysis of variance (for normally distributed data) or the Kruskal‐Wallis test (for nonnormally distributed data).

^a The value is significantly different from no or mild PH (P < .05).

A total of 41 dogs (76%) with MMVD and detectable TR died within the follow‐up time, including 32 cardiac‐related deaths (59%) and 9 non‐cardiac‐related deaths (16%): 21/33 (64%) dogs with no or mild PH, 11/11 (100%) dogs with moderate PH, and 9/10 (90%) dogs with severe PH. Thirteen dogs were right‐censored (9 alive, 4 lost to follow‐up). The median follow‐up time of this study was 579 days (95% CI: 448‐709). The median survival times for all‐cause mortality and cardiac‐related death of dogs with MMVD and detectable TR were 433 days (95% CI: 331‐536) and 408 days (95% CI: 292‐523), respectively.

3.2. Echocardiographic measurements

The results of the comparison of echocardiographic variables based on PH severity are summarized in Table 2. For left‐heart morphology and function, dogs with severe PH had significantly greater E and E/e′ than those with no or mild PH. There were no significant associations between PH severity and LV size (LVIDDN and LVIDSN). The LA/Ao was significantly higher in dogs with moderate PH compared with that with no or mild PH. For RV morphology, significantly higher RV size (RVEDA index, RVESA index, and RVIDd) and significantly lower PV VTI, RV‐SL, and RV‐SrL_6seg were observed in dogs with severe PH compared with those with the other PH severity groups. Additionally, PVRecho was significantly higher with the progression of PH severity. Two dogs with moderate PH and RHF had high values of PVRecho (both PVRecho were 2.02 and 2.91).

TABLE 2.

Echocardiographic variables comparing 54 dogs with myxomatous mitral valve disease and detectable tricuspid regurgitation based on pulmonary hypertension severity.

Variables	PH severity (n = 54)			P *
	No or mild (n = 33)	Moderate (n = 11)	Severe (n = 10)
LA/Ao	1.7 (1.3‐2.2)	2.3 (1.9‐2.4) a	1.9 (1.7‐2.4)	.02
LVIDDN (cm/kg0.294)	1.8 (1.5‐2.0)	2.0 (1.5‐2.5)	1.8 (1.1‐2.3)	.59
LVIDSN (cm/kg0.315)	1.0 (0.8‐1.1)	0.9 (0.8‐1.3)	0.9 (0.6‐1.1)	.48
Fractional shortening (%)	44.3 (39.5‐48.2)	44.6 (39.2‐52.9)	48.7 (41.6‐57.1)	.17
E (m/s)	1.0 (0.8‐1.2)	1.3 (1.1‐1.4) a	1.3 (0.7‐1.6) a	.04
E/A	1.1 (0.9‐1.5)	1.3 (1.1‐2.2)	1.2 (0.6‐1.9)	.38
E/e'	13.5 (11.3‐16.9)	15.4 (12.5‐18.1)	20.5 (15.8‐22.9) a	.03
RVEDA index (cm2/kg0.624)	0.9 (0.7‐1.0)	1.0 (0.8‐1.2)	1.3 (1.1‐1.7) a , b	<.01
RVESA index (cm2/kg0.628)	0.4 (0.4‐0.5)	0.5 (0.3‐0.6)	0.7 (0.6‐0.9) a , b	<.01
RVIDd index (mm/kg0.327)	5.9 (5.3‐7.3)	7.3 (5.5‐8.1)	10.1 (7.5‐12.0) a , b	<.01
TAPSEn (mm/kg0.284)	7.4 (6.6‐8.7)	8.0 (7.3‐9.1)	7.2 (5.2‐8.6)	.36
RV FACn (%/kg−0.097)	58.8 (49.8‐62.9)	58.7 (49.3‐61.2)	49.2 (36.9‐56.9) a	.03
RV s' (cm/s)	12.1 (9.6‐13.6)	12.0 (10.5‐16.7)	11.8 (7.2‐16.0)	.79
PV VTI (cm)	8.5 (6.7‐9.6)	7.3 (6.1‐10.2)	6.8 (5.1‐7.8) a , b	<.01
PVRecho	1.1 (0.7‐1.6)	1.8 (1.6‐2.2) a	3.7 (2.9‐5.0) a , b	<.01
RV‐SL3seg (%)	28.7 (23.9‐31.6)	28.1 (25.1‐30.1)	22.3 (16.8‐25.2) a , b	.03
RV‐SrL3seg (%/s)	5.1 (3.3‐7.2)	4.9 (3.3‐6.7)	3.1 (2.2‐5.7)	.28
RV‐SL6seg (%)	25.6 (20.1‐29.3)	24.5 (22.5‐28.3)	16.3 (13.3‐22.6) a , b	<.01
RV‐SrL6seg (%/s)	3.5 (2.7‐4.7)	3.1 (2.3‐3.8)	2.4 (1.5‐3.5) a	<.01

Note: Continuous data is expressed as median (interquartile range).

Abbreviations: 3seg, only RV free wall analysis; 6seg, global RV analysis; E, early‐diastolic transmitral flow velocity; E/A, early‐diastolic to late diastolic transmitral flow velocity ratio; e', tissue Doppler imaging‐derived peak early‐diastolic myocardial velocity of the septal mitral annulus; LA/Ao, left atrium to aortic diameter ratio; LVIDDN, end‐diastolic left ventricular internal dimension normalized by body weight; LVIDSN, end‐systolic left ventricular internal dimension normalized by body weight; PV VTI, velocity‐time integral of the pulmonary artery flow; PVRecho, pulmonary vascular resistance estimated by echocardiography; RV FACn, RV fractional area change normalized by body weight; RV s', tissue Doppler imaging‐derived peak systolic myocardial velocity of the lateral tricuspid annulus; RV, right ventricular; RVEDA index, end‐diastolic RV area normalized by body weight; RVESA index, end‐systolic RV area normalized by body weight; RVIDd index, end‐diastolic RV internal dimension normalized by body weight; RV‐SL, RV longitudinal strain; RV‐SrL, RV longitudinal strain rate; TAPSEn, tricuspid annular plane systolic excursion normalized by body weight; TR, tricuspid regurgitation.

^* P‐values of 1‐way analysis of variance (for normally distributed data) or the Kruskal‐Wallis test (for nonnormally distributed data).

^a The value is significantly different from no or mild PH (P < .05).

^b The value is significantly different from moderate PH (P < .05).

The first and second tertile values of PVRecho in 54 dogs with MMVD and detectable TR were 1.11 and 1.77, respectively. Echocardiographic data classified by PVRecho tertiles are presented in Table S2.

3.3. Survival analysis

In the univariable Cox proportional hazard analysis of the prognostic value of echocardiographic variables in dogs with MMVD, the LA/Ao ratio, LVIDDN, LVIDSN, E, E/A, RVEDA index, RVESA index, RVIDd index, PV VTI, TR velocity, and PVRecho were significantly different (Table 3). No significant associations were observed in any RV functional variables, including TAPSEn (P = .14), RV FACn (P = .57), RV s' (P = .06), speckle tracking echocardiography‐derived RV‐SL (RV‐SL_3seg, P = .80; RV‐SL_6seg, P = .47) and RV‐SrL (RV‐SrL_3seg, P = .30; RV‐SrL_6seg, P = .88). After adjusting for age, sildenafil administration, and ACVIM stage, the LA/Ao ratio, LVIDDN, E, RVIDd index, and PVRecho were included in the multivariable model; the LA/Ao ratio and PVRecho remained significant in the multivariable model (adjusted multivariable HR [95% CI]: LA/Ao, 1.2 [1.1‐1.3]; PVRecho, 2.1 [1.6‐3.0]).

TABLE 3.

Univariable and multivariable cox proportional hazard analysis for echocardiographic variables in 54 dogs with myxomatous mitral valve disease and detectable tricuspid regurgitation.

Variables	Univariable analysis		Multivariable analysis a
	HR (95% CI)	P	P
LA/Ao (per 0.1 increase)	1.10 (1.05‐1.15)	<.01	<.01
LVIDDN (per 0.1 cm/kg0.294 increase)	1.16 (1.05‐1.29)	<.01
LVIDSN (per 0.1 cm/kg0.315 increase)	1.16 (1.00‐1.34)	.05
E (per 0.1 m/s increase)	1.15 (1.04‐1.28)	<.01
E/A (per 0.1 increase)	1.09 (1.04‐1.14)	<.01
RVEDA index (per 0.1 cm2/kg0.624 increase)	1.16 (1.05‐1.29)	<.01
RVESA index (per 0.1 cm2/kg0.628 increase)	1.31 (1.06‐1.62)	.01
RVIDd index (per 0.1 mm/kg0.327 increase)	1.30 (1.14‐1.49)	<.01
RV s' (per 1.0 cm/s decrease)	0.91 (0.82‐1.01)	.06
PV VTI (per 1.0 decrease)	1.29 (1.08‐1.56)	<.01
TR velocity (per 0.1 m/s increase)	1.10 (1.05‐1.15)	<.01
PVRecho (per 1.0 increase)	1.95 (1.50‐2.55)	<.01	<.01

Abbreviations: CI, confidence interval; E, early‐diastolic transmitral flow velocity; E/A, early‐diastolic to late diastolic transmitral flow velocity ratio; HR, hazard ratio; LA/Ao, left atrium to aortic diameter ratio; LVIDDN, end‐diastolic left ventricular internal dimension normalized by body weight; LVIDSN, end‐systolic left ventricular internal dimension normalized by body weight; PVRecho, pulmonary vascular resistance estimated by echocardiography; PV VTI, velocity‐time integral of the pulmonary artery flow; RV, right ventricular; RVEDA index, end‐diastolic RV area normalized by body weight; RVESA index, end‐systolic RV area normalized by body weight; RVIDd index, end‐diastolic RV internal dimension normalized by body weight; RV s', tissue Doppler imaging‐derived peak systolic myocardial velocity of the lateral tricuspid annulus; TR, tricuspid regurgitation.

^a The multivariable model was adjusted for age, sildenafil administration, and American College of Veterinary Internal Medicine stage.

Figure 3 shows the results of the Kaplan‐Meier curves adjusted for age, sildenafil administration, and ACVIM stage, and log‐rank tests used to evaluate the influence of PVRecho on all‐cause mortality and cardiac‐related death. Dogs with PVRecho >1.77 showed significant associations with low survival rates in all‐cause mortality compared with those with PVRecho ≤1.77 (the median survival times [95% CI]: PVRecho <1.11, 984 days [567‐incalculable]; PVRecho 1.11‐1.77, 662 days [178‐968]; PVRecho >1.77, 378 days [93‐495]; PVRecho <1.11 vs PVRecho >1.77, P < .01; PVRecho 1.11‐1.77 vs PVRecho >1.77, P = .04). There were no significant differences in survival rates between dogs with PVRecho <1.11 and those PVRecho 1.11‐1.77 (P = .30; Figure 3A). For cardiac‐related death, there were significant differences in survival rates among dogs with PVRecho <1.11, 1.11‐1.77, and >1.77 (the median survival times [95% CI]: PVRecho <1.11, not reached; PVRecho 1.11‐1.77, 549 days [167‐991]; PVRecho >1.77, 378 days [93‐495]; PVRecho <1.11 vs PVRecho 1.11‐1.77, P = .02; PVRecho <1.11 vs PVRecho >1.77, P < .01; PVRecho 1.11‐1.77 vs PVRecho >1.77, P = .04; Figure 3B).

FIGURE 3.

Kaplan‐Meier survival curves showing the significant effects of PVRecho on the all‐cause mortality (log‐rank test: P < .01, A) and cardiac‐related death (log‐rank test: P < .01, B) of dogs with myxomatous mitral valve disease and tricuspid regurgitation. Each curve was adjusted for age, sildenafil administration, and American College of Veterinary Internal Medicine stage. Black, green, and blue lines represent the survival curves of dogs with PVRecho <1.11, 1.11‐1.77, and >1.77, respectively. PVRecho, pulmonary vascular resistance estimated by echocardiography.

4. DISCUSSION

This study revealed that dogs with severe PH based on TR velocity had lower RV‐SL and increased RV dimensions. Additionally, PVRecho was significantly higher with the progression of PH severity. These results suggest that dogs with severe PH might have RV dysfunction and dilatation associated with increased RV pressure overload, and that increased PVRecho might be associated with RV maladaptation and the progression to Cpc‐PH. In addition, Cox proportional hazard analysis showed that an increased LA/Ao ratio and PVRecho were independent prognostic factors in dogs with MMVD. Our results indicate that PVRecho and LA/Ao ratio could provide prognostic information in dogs with MMVD and detectable TR.

This study evaluates the prognostic value of PVRecho in dogs with MMVD and PH. In addition to the increased LA/Ao ratio which has been reported as the prognostic indicator in dogs with MMVD, ³ , ¹⁶ , ³⁸ , ³⁹ , ⁴⁰ multivariable analysis showed that an increased PVRecho was an independent prognostic factor in dogs with PH secondary to MMVD. Especially, in dogs with more than the second tertile of PVRecho (ie, PVRecho >1.77), median survival time was reduced to approximately 1 year. Previous human studies have reported the prognostic importance of PVRecho in patients with PH secondary to interstitial lung disease ¹³ , ⁴¹ and stable coronary artery disease. ⁴² Additionally, some human studies have reported that progression to Cpc‐PH in patients with left heart disease is associated with poor prognosis. ⁴ , ⁵ , ⁶ , ⁷ , ⁸ The results of this study suggest that increased PVRecho could reflect increased PVR (ie, progression to Cpc‐PH) and poor prognosis also in dogs with MMVD and detectable TR. Consequently, the PVRecho assessment may also provide important long‐term prognostic information in dogs with MMVD and detectable TR. However, this study used PVRecho at a single point in time.

In this study, the PH severity was clinically classified according to the TR velocity. In our study samples, almost all dogs with severe PH had MMVD progressed to the point of developing left‐sided congestive heart failure (ie, ACVIM Stage C/D). This indicates that in addition to the increased left atrial pressure, hypoxia because of left‐sided congestive heart failure might have induced pulmonary vascular remodeling and more advanced PH. Therefore, our results suggest that a comprehensive cardiac evaluation including the right heart should be performed especially in dogs with ACVIM Stage C/D. However, 4 dogs with MMVD had active CHF at the time of study inclusion, and the resulting hypoxia and neurohormonal activation might worsen PH temporarily and influence the results. ⁴³

Significantly higher RV size indicators and lower RV functional indicators were observed in dogs with severe PH based on TR velocity. However, 2 dogs were diagnosed with RHF despite being classified as having moderate PH from TR velocity. Although a TR pressure gradient >55 mm Hg was reported to be an indicator for poor prognosis in dogs with MMVD, ³ another study reported that determining pulmonary artery pressure based on TR velocity might cause a measurement error of up to 20 mm Hg from that measured invasively. ⁴⁴ Furthermore, severely depressed RV function and low‐volume loading conditions might underestimate pulmonary artery pressure determinations based on the TR velocity. ¹ , ⁴⁵ Therefore, the recent ACVIM consensus for the diagnosis of PH in dogs was concerned about a high risk of misdiagnosing PH using TR velocity alone and advocated a comprehensive diagnosis that includes not only TR velocity but also right‐heart morphology and function. ¹ Our results also indicated that the evaluation of PH pathophysiology by TR velocity alone can be misleading. On the other hand, a significantly increased PVRecho was observed in dogs with severe PH. Especially, 2 dogs with RHF had severely high PVRecho despite having moderate PH based on TR velocity. Because PVRecho is calculated using TR velocity, the indicator of RV afterload, and PV VTI, the indicator of RV function, the PVRecho variable reflects the balance between RV systolic function and loading conditions (ie, RV performance and RV‐pulmonary arterial coupling) in addition to the increased PVR. ¹⁴ This suggests that increased PVRecho is associated with reduced RV performance, which might be related to progression to Cpc‐PH in dogs with MMVD and PH. Overall, our results indicate that PVRecho might also be an additional tool that could reflect RV performance and stratify the PH pathophysiology in dogs with MMVD and detectable TR.

Although a significant association between RV dilatation and survival was observed in this study, no RV functional variables, including speckle‐tracking echocardiography variables, showed a significant association with survival. The main factor leading to this result might be the adaptation mechanism of the RV against afterload. Naturally, the RV is sensitive to afterloads; when the RV afterload increases, the RV will become hypertrophied and hyperactivated to maintain cardiac output. ³⁵ However, an excessively elevated RV afterload may cause RV dysfunction and result in RV dilatation. ³⁵ Therefore, unlike the ever‐increasing variables associated with PH progression, such as PVRecho and RV size indicators, a temporary increase in RV function against PH might make it difficult to predict long‐term prognosis using RV functional indicators in dogs with PH. In addition, various factors might overestimate the RV function based on echocardiographic variables such as TAPSE, RV FAC, and RV s'; these include heart rate, volume‐loading conditions, severity of TR, and ventricular interdependence. ¹⁹ , ²³ , ⁴⁶ These factors might also hinder the detection of RV dysfunction in dogs with post‐capillary PH. However, because there was a relatively small population of dogs with progressed PH and RHF that were likely to show RV dysfunction, ¹⁸ , ¹⁹ , ⁴⁷ our results might have been different if the study included dogs with more severe PH and RHF.

This study had several limitations. First, this study could not perform right‐heart catheterization. Therefore, the actual PVR values, and whether the isolated post‐capillary PH progressed to Cpc‐PH, are unknown. Second, we were unable to standardize medications. In particular, a pulmonary vasodilator (sildenafil) might influence the results of PVRecho and mortality. Third, not all dogs received a complete diagnostic evaluation to screen for comorbid diseases contributing to PH. Therefore, diseases other than MMVD (ie, pre‐capillary PH) might have contributed to increased pulmonary arterial pressure and/or PVR. ¹ Finally, the small sample size, especially for dogs with severe PH and/or RHF, may have influenced our results.

In conclusion, a significantly higher value for PVRecho was observed in dogs with severe PH secondary to MMVD as compared to those with mild and moderate PH, possibly reflecting the imbalance of RV systolic function and loading conditions and the progression to Cpc‐PH. Moreover, Cox proportional hazard analyses revealed that increases in the LA/Ao ratio and PVRecho could provide prognostic information for dogs with MMVD and PH. These results support the prognostic utility of measuring PVRecho in dogs with MMVD and PH. Further studies that include more severe cases are warranted to further explore prognostic prediction in dogs with MMVD.

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

All procedures followed the Guidelines for IACUC of Nippon Veterinary and Life Science University in Japan, and the study was approved by the Ethical Committee for Animal Use of Nippon Veterinary and Life Science University Veterinary Medical Teaching Hospital, Japan (approval number: R2‐5).

HUMAN ETHICS APPROVAL DECLARATION

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

Supporting information

Table S1. Clinical data classified by tertiles of pulmonary vascular resistance estimated by echocardiography in 54 dogs with myxomatous mitral valve disease and detectable tricuspid regurgitation.

Table S2. Echocardiographic data classified by tertiles of pulmonary vascular resistance estimated by echocardiography in 54 dogs with myxomatous mitral valve disease and detectable tricuspid regurgitation.

Yuchi Y, Suzuki R, Yasumura Y, et al. Prognostic value of pulmonary vascular resistance estimated by echocardiography in dogs with myxomatous mitral valve disease and pulmonary hypertension. J Vet Intern Med. 2023;37(3):856‐865. doi: 10.1111/jvim.16686