Reduced risk of arterial thromboembolism in cats with pleural effusion due to congestive heart failure

Abstract

Objectives

The aim of the study was to determine whether cardiogenic pleural effusion in cats is associated with a lower risk of arterial thromboembolism (ATE) compared with cats with cardiac disease without evidence of pleural effusion.

Methods

A cross-sectional study was conducted on owned cats with natural occurring cardiac diseases. Cats included were classified in three groups: those with cardiac disease but no evidence of congestive heart failure (CHF); those with evidence of cardiogenic pulmonary oedema; and those with evidence of cardiogenic pleural effusion. Prevalence of ATE was calculated and the variables analysed for an association with this outcome were the presence and type of CHF, sex and neuter status, age, breed, type of cardiac diseases and left atrial (LA) dimension. A multivariable logistic regression model was used to fit the association between ATE and these variables.

Results

A total of 366 cats with cardiac disease met the inclusion criteria: 179 were included in the group with cardiac disease but no evidence of CHF, 66 in the group with evidence of cardiogenic pulmonary oedema and 121 in the group with evidence of cardiogenic pleural effusion. Prevalence of ATE (58/366 [15.8%]) was significantly different among groups (with no evidence of CHF, 28/179 [15.6%]; with evidence of cardiogenic pulmonary oedema, 22/66 [33.3%]; with evidence of cardiogenic pleural effusion, 8/121 [6.6%]; P <0.001). Cats with ATE had a significantly higher LA to aortic root ratio (2.30 ± 0.46) than those without ATE (2.04 ± 0.46; P <0.001). Multivariable logistic regression analysis indicated that the group with evidence of cardiogenic pleural effusion was associated with a lower risk of developing ATE compared with groups with cardiac disease but no evidence of CHF and with evidence of cardiogenic pulmonary oedema (P = 0.005 and P <0.001, respectively).

Conclusions and relevance

Presence of cardiogenic pleural effusion is associated with a lower risk of developing ATE, while LA enlargement is a risk factor for ATE.

Keywords

Fibrinolysis oedema thrombosis heart disease

Introduction

Arterial thromboembolism (ATE) is a common complication of cardiomyopathy in cats;^1–3 on presentation, in cats with hypertrophic cardiomyopathy (HCM), it is the second most common cause of clinical signs after congestive heart failure (CHF).^4,5

Thromboemboli are believed to originate from fragmentation or dislodgment of left atrial (LA) or LA appendage thrombi.⁶ Although the pathogenesis of LA thrombus formation in cats is unresolved, it is assumed that thrombi form because of endothelial damage,⁷ activation of platelets and coagulation factors causing a procoagulant imbalance,^8–10 and local stagnation of blood flow due to poor LA contractility and enlargement (Virchow’s triad).¹¹

In dogs, horses and humans, abdominal and pleural effusions have been shown to have intrinsic fibrinolytic activity due to the presence of all the proteins that participate in coagulation and fibrinolysis^12–17 and due to the activities of mesothelial cells that further increase anticoagulation.^18–21 Dogs with pleural and abdominal effusions show enhanced systemic fibrinolysis,^22–24 probably due to reabsorption of these pathological effusions.²⁵ This hyperfibrinolytic state may increase the risk of bleeding, reducing the risk of thrombosis.²⁶

We therefore hypothesised that cats with cardiogenic pleural effusion were less likely to present with ATE compared with cats with cardiac disease without effusions. Therefore, the aim of this study was to determine whether the presence of cardiogenic pleural effusion in cats is associated with a decreased risk of developing ATE in comparison to cats with cardiogenic pulmonary oedema or with non-congestive cardiac disease.

Materials and methods

Study population and study design

A cross-sectional study was conducted retrospectively, evaluating the prevalence of ATE in client-owned cats with cardiac disease that presented with a heart murmur, gallop rhythm or suspected cardiac disease at San Marco Veterinary Clinic between 1 June 2005 and 31 December 2018. Records were retrieved from the electronic medical database, P.O.A System-Plus9.0, by searching for echocardiographic reports in cats. The inclusion criteria were: the presence of a complete medical record (patient signalment, history and full physical examination findings); having undergone thoracic radiography (if their clinical condition allowed this procedure safely); and all cats had an echocardiography performed for the diagnosis of their cardiac disease with the report available for review. Cats were excluded if the LA to aortic root ratio (LA:Ao) had not been reported. Cats with an LA:Ao <1.5 were also excluded to increase the specificity of cardiogenic ATE in the studied population. Only data from the first presentation for cardiac disease and diagnosis of ATE were included.

Group definitions and diagnosis of ATE

Cats were divided into three groups according to presence and type of CHF: cats with cardiac diseases but no evidence of CHF (group 1); cats with evidence of cardiogenic pulmonary oedema but no pleural effusion (group 2); and cats with evidence of cardiogenic pleural effusion with or without cardiogenic pulmonary oedema (group 3).

Pulmonary oedema and pleural effusion were considered secondary to CHF when clinical history and physical examination findings, thoracic radiographs and ultrasonographic evidence of structural or functional cardiac disease were consistent with this complex clinical syndrome.²⁷ In cats with CHF, thoracic radiographs were taken in most cases as their clinical condition allowed this procedure to be carried out safely or point of care thoracic ultrasonography did not clarify the cause of the respiratory signs. In cats with pleural effusion, thoracocentesis was performed for diagnostic or therapeutic purposes when judged necessary by the attending clinician.

A diagnosis of ATE was based on a combination of compatible clinical findings (eg, absence of femoral pulse, cold hindlimbs, pallor or cyanosis of the nail-beds, paraparesis/paralysis and severe pain) and, in most cases, compatible abdominal Doppler ultrasonography results (ie, absence of blood flow and/or identification of a visible thrombus in the terminal aorta).^28,29 Based on this diagnosis, cats were also divided into those with and those without ATE.

Echocardiography

The diagnosis of cardiac disease was determined by diagnostic-quality two-dimensional, colour flow Doppler and M-mode echocardiographic examinations. LA:Ao was obtained from a right parasternal short-axis view at the level of the heart base. It was calculated from the ratio of LA diameter to aortic root diameter measured at the first diastolic frame in which closure of the aortic valve leaflets was evident.³⁰

For the purpose of this study, cardiac diseases were classified as follows: hypertrophic cardiomyopathy (HCM); restrictive cardiomyopathy (RCM); dilated cardiomyopathy; non-specific cardiomyopathy phenotype (including all other cardiomyopathies not adequately falling into the previous categories); and other cardiac diseases (mainly including acquired valvular diseases or congenital cardiac defects).

Ethical approval

All diagnostic and therapeutic procedures reported in this non-experimental study were performed by the attending clinician solely for the cat’s benefit with prior informed written consent from the owner. Anaesthesia, euthanasia or experimental procedures were not required for any part of the study. All the procedures performed complied with the European legislation ‘on the protection of animals used for scientific purposes’ (Directive 2010/63/EU) and with the ethical requirements of Italian law (Decreto Legislativo 04/03/2014, n. 26). All data were retrospectively retrieved from the medical records of the cats enrolled in the study and no personal identifiable information was used. Accordingly, this type of study did not require an authorisation or an ID protocol number from an institutional animal care and use committee.

Statistical analysis

Continuous data were assessed for normality of distribution with the Shapiro–Wilk test. Normally distributed data are reported as a mean ± SD and non-normally distributed data are reported as a median and interquartile range (IQR).

Differences between possible confounders such as sex and neuter status, breed and type of cardiac disease (χ² test) and age (Kruskal–Wallis followed by Mann–Whitney test) were evaluated among and between the three groups. Similarly, differences between breed and sex and neuter status (χ² test), age (Mann–Whitney test) and type of cardiac disease (χ² test with Monte Carlo simulation) were also evaluated between cats with ATE and those without ATE.

To determine whether cats with evidence of cardiogenic pleural effusion had a lower occurrence of ATE in comparison with those with evidence of cardiogenic pulmonary oedema or with cardiac disease but no evidence of CHF, the prevalence of ATE among and between the three groups was compared using the χ² test. The LA:Ao among and between the three groups was compared by ANOVA, followed by the Tamhane post-hoc test. Then, to determine whether LA enlargement was a risk factor for ATE differences in LA:Ao between cats with ATE and those without ATE, a t-test was used.

Finally, a multivariable logistic regression model was used to fit the association between the outcome variable (ie, ATE) and the independent predictors previosly univariately analysed, with a significant P value of <0.05, after evaluation of autocorrelation and interaction between predictors. The Exp(B) was calculated and reported as odds ratios (ORs) with 95% confidence intervals (CIs) for each incremental unit of the continuous variables or for the level of interest of the categorical variables, respectively.

For all statistical analyses, the significance level was set to α = 0.05.

Results

Study population characteristics

A search of the San Marco Veterinary Clinic database showed that during the study period, 4060 cats presented to the clinic and 641 cats with cardiac diseases were provisionally eligible for the study. Of these cases, 65% were first-opinion and 35% were referral cases. None of the cats were on anti-aggregant treatment and <15% of the referral cats were on drugs for their cardiac disease. Twelve cats were subsequently excluded from the study: one because it had Cor triatriatum sinister, which prevented LA measurement; and 11 due to LA:Ao measurements missing from the echocardiographic reports. Moreover, a further 263 cats were excluded because they had an LA:Ao <1.5, leaving 366 cats for the final analysis. Only three of the excluded cats had been diagnosed with ATE (Figure 1).

Figure 1

Flow chart of the study population selection. Group 1: cats with cardiac disease but no evidence of congestive heart failure; group 2: cats with evidence of cardiogenic pulmonary oedema; and group 3: cats with evidence of cardiogenic pleural effusion. ATE = arterial thromboembolism; LA:Ao = left atrial to aortic root ratio

ATE was detected at presentation in 58 of the 366 cats, with an overall estimated prevalence of 15.8%. A total of 179 cats with cardiac disease did not have evidence of CHF, 66 had evidence of cardiogenic pulmonary oedema but no pleural effusion and 121 had evidence of cardiogenic pleural effusion (with or without concurrent cardiogenic pulmonary oedema; two of these cats also presented with ascites) (Figure 1). Of the 121 cats with evidence of cardiogenic pleural effusion, 61 underwent diagnostic or therapeutic thoracocentesis. In all cases, cytological and biochemical analyses, including serum and pleural effusion lactate dehydrogenase concentrations,^31,32 were consistent with transudates caused by increased hydrostatic pressure, consistent with a cardiac origin.

Of the cats, 225 were male (27 intact and 198 neutered) and 141 were female (21 intact and 120 neutered). There was an overall significant difference in sex and neuter status among the three groups (χ² = 20.37; P = 0.002). Cats with cardiac disease but no evidence of CHF and cats with evidence of cardiogenic pulmonary oedema contained more male neutered and fewer female neutered cats compared with cats with evidence of cardiogenic pleural effusion (Table 1). There were more male neutered and fewer female neutered cats among cats presented with ATE compared with cats presented without ATE (χ² = 7.84; P = 0.049) (Table 1).

Table 1

Sex and neuter status distribution of cats without evidence of congestive heart failure (CHF), cats with cardiogenic pulmonary oedema and cats with cardiogenic pleural effusion, and in cats with and without arterial thromboembolism (ATE)

	Neuter status				χ²	P value	Comparisons between groups
	ME	MN	FE	FN	χ²	P value	Comparisons between groups
Group
1 (without CHF) (n = 179)	17 (9.5)	105 (58.7)	6 (3.4)	51 (28.5)	20.37	0.002	G₁ vs G₂ χ² = 4.50P = 0.212
2 (pulmonary oedema) (n = 66)	5 (7.6)	41 (62.1)	6 (9.1)	14 (21.2)			G₁ vs G₃ χ² = 14.52P = 0.002
3 (pleural effusion) (n = 121)	5 (4.1)	52 (43.5)	9 (7.4)	55 (45.5)			G₂ vs G₃ χ² = 11.04P = 0.011
Presence of ATE
With ATE (n = 58)	2 (3.4)	41 (70.7)	2 (3.4)	13 (22.4)	7.84	0.049
Without ATE (n = 308)	25 (8.1)	157 (51.0)	19 (6.2)	107 (34.7)	7.84	0.049

Data are presented as n (%)

FE = female entire; FN = female neutered; G₁ = group 1; G₂ = group 2; G₃ = group 3; ME = male entire; MN = male neutered

Median age at diagnosis was 108 months (IQR 62–153 months). There was an overall significant difference in age within the three groups (H = 19.59; P <0.001), with cats with evidence of cardiogenic pulmonary oedema being the youngest and cats with evidence of cardiogenic pleural effusion the oldest (Table 2). There was no difference in age in cats presented with or without ATE (U = 8678; P = 0.732) (Table 2).

Table 2

Age distribution of cats without evidence of congestive heart failure (CHF), cats with cardiogenic pulmonary oedema and cats with cardiogenic pleural effusion and in cats with and without arterial thromboembolism (ATE)

	Age (months)		P value	Comparisons between groups
Group		Kruskal–Wallis
1 (without CHF) (n = 179)	100 (55–150)	H = 19.59	<0.001	G₁ vs G₂ U = 4865P = 0.034
2 (pulmonary oedema) (n = 66)	84 (37–114)			G₁ vs G₃ U = 12964P = 0.004
3 (pleural effusion) (n = 121)	126 (85–163)			G₂ vs G₃ U = 5533P <0.001
Presence of ATE
With ATE (n = 58)	108 (73–137)	Mann–Whitney U = 8678	0.732
Without ATE (n = 308)	112 (64–157)	Mann–Whitney U = 8678	0.732

Data are presented as median (IQR)

G1 = group 1; G2 = group 2; G3 = group 3; IQR = interquartile range

A total of 98 cats were purebreds, including 31 Persian cats, 15 Sphynx, eight Birmans, eight Chartreux, eight Siamese, six Maine Coons, six British Shorthairs, five Exotic Shorthairs, three Scottish Folds and three Norwegian Forest Cats. The remaining breeds (Burmese, Bombay, Siberian, Devon Rex and Oriental Shorthair) were represented by a single cat. The remaining 268 cats were domestic shorthairs or longhairs. There was no difference in breeds within the three groups (χ² = 6.8; P = 0.339) (Table 3). There was no difference in breeds between cats presented with or without ATE (χ² = 1.40; P = 0.705) (Table 3).

Table 3

Breed distribution of cats without evidence of congestive heart failure (CHF), cats with cardiogenic pulmonary oedema and cats with cardiogenic pleural effusion

	Breed				χ²	P value
	DSH/DLH	Persian	Sphynx	Others	χ²	P value
Group
1 (without CHF) (n = 179)	124 (69.3)	20 (11.2)	9 (5.0)	26 (14.5)	6.81	0.339
2 (pulmonary oedema) (n = 66)	50 (75.8)	4 (6.1)	4 (6.1)	8 (12.1)
3 (pleural effusion) (n = 121)	94 (77.7)	7 (5.8)	2 (1.7)	18 (14.9)
Presence of ATE
With ATE (n = 58)	42 (72.4)	5 (8.6)	1 (1.7)	10 (17.2)	1.40	0.705
Without ATE (n = 308)	226 (73.4)	26 (8.4)	14 (4.5)	42 (13.6)	1.40	0.705

Data are presented as n (%)

DLH = domestic longhair; DSH = domestic shorthair; ATE = arterial thromboembolism

A total of 220 cats were diagnosed with HCM, 54 with RCM, 29 with dilated cardiomyopathy and 26 with non-specific cardiomyopathy phenotype; 37 were classified as ‘other cardiac diseases’. There was an overall significant difference in the types of cardiac disease identified among the three groups (χ² = 47.38; P <0.001) with more cats with HCM in the group of cats with cardiac disease but no evidence of CHF compared with the other two groups, and more cats with RCM and dilated cardiomyopathy in the group of cats with evidence of cardiogenic pleural effusion compared with the other two groups, respectively (Table 4). Similarly, there was an overall significant difference in the types of cardiac disease in cats presented with and without ATE (χ² = 11.79; P <0.019) with more cats with HCM and fewer with ‘other cardiac diseases’ in those with ATE.

Table 4

Frequency of type of cardiac diseases in cats without evidence of congestive heart failure (CHF), cats with cardiogenic pulmonary oedema and cats with cardiogenic pleural effusion and in cats with and without arterial thromboembolism (ATE)

	Type of cardiac diseases					χ²	P value	Comparisons between groups
	HCM	RCM	DCM	Non-specific cardiomyopathy phenotype	Others	χ²	P value	Comparisons between groups
Group
1 (without CHF) (n = 179)	130 (72.6)	15 (8.4)	7 (3.9)	6 (3.4)	21 (11.7)	47.38	<0.001	G₁ vs G₂ χ² = 4.39P = 0.356
2 (pulmonary oedema) (n = 66)	42 (63.6)	10 (15.2)	4 (6.1)	4 (6.1)	6 (9.1)			G₁ vs G₃ χ² = 46.03P <0.001
3 (pleural effusion) (n = 121)	48 (39.7)	29 (24.0)	18 (14.9)	16 (13.2)	10 (8.3)			G₂ vs G₃ χ² = 11.59P = 0.021
Presence of ATE
With ATE (n = 58)	44 (75.9)	9 (15.5)	3 (5.2)	2 (3.4)	0 (0)	11.79	0.019
Without ATE (n = 308)	176 (57.1)	45 (14.6)	26 (8.4)	24 (7.8)	37 (12.0)	11.79	0.019

Data are presented as n (%)

DCM = dilatated cardiomyopathy; G₁ = group 1; G₂ = group 2; G₃ = group 3; HCM = hypertrophic cardiomyopathy; OR = odds ratio; RCM = restrictive cardiomyopathy

Prevalence of ATE among groups and differences in LA size among groups and between cats with and without ATE

The estimated prevalence of ATE in cats with no evidence of CHF was 15.6% (28/179); in cats with evidence of cardiogenic pulmonary oedema it was 33.3% (22/66); and in cats with evidence of cardiogenic pleural effusion it was 6.6% (8/121). This was significantly different among the three groups (χ² = 22.88; P <0.001), with cats with evidence of cardiogenic pleural effusion having significantly less ATE events compared with the other two groups (Table 5).

Table 5

Prevalence of arterial thromboembolism (ATE) in cats without evidence of congestive heart failure (CHF), cats with cardiogenic pulmonary oedema and cats with cardiogenic pleural effusion

Group	ATE	χ²	P value	Comparisons between groups
1 (without CHF) (n = 179)	28 (15.6)	22.88	<0.001	G₂ vs G₁ P = 0.004OR = 2.70 (95% CI 1.34–5.44)
2 (pulmonary oedema) (n = 66)	22 (33.3)			G₁ vs G₃ P = 0.029OR = 2.62 (95% CI 1.09–6.50)
3 (pleural effusion) (n = 121)	8 (6.6)			G₂ > G₃ P <0.001OR = 7.06 (95% CI 2.73–18.80)

Data are presented as n (%)

CI = confidence interval; G₁ = group 1; G₂ = group 2; G₃ = group 3; OR = odds ratio

LA:Ao was significantly different among the three groups (F = 30.87; P <0.001) (Figure 2). LA:Ao was significantly bigger in cats with evidence of cardiogenic pulmonary oedema and pleural effusion compared with cats with cardiac disease but no evidence of CHF (P <0.001 for both comparisons), but there was no significant difference in LA:Ao between cats with evidence of cardiogenic pulmonary oedema and pleural effusion (P = 0.349).

Figure 2

Boxplot of left atrial to aortic root ratio (LA:Ao) in cats with cardiac disease but no evidence of congestive heart failure (group 1), cats with evidence of cardiogenic pulmonary oedema (group 2) and cats with evidence of cardiogenic pleural effusion (group 3). LA:Ao: group 1 = 1.90 ± 0.41; group 2 = 2.32 ± 0.49; group 3 = 2.21 ± 0.43. (a,b) Data with different letters are significantly different (P <0.001 for both comparisons). The bottom and top of the box are the first and third quartiles; the band within the box is the median. The whiskers correspond to the lowest datum still within 1.5 interquartile ranges (IQRs) of the first quartile, and the hightest datum still within 1.5 IQRs of the fourth quartile. Circles are outlier values (>1.5 IQRs away from the closest end of the box). Asterisks are extreme outlier values (>3 IQRs away from the closest end of the box)

LA:Ao was significantly greater in cats with ATE than in those without ATE (t = 3.99; P <0.001) (Figure 3).

Figure 3

Boxplot of left atrial to aortic root ratio (LA:Ao) in cats without arterial thromboembolism (ATE) and cats with ATE. LA:Ao: cats with ATE = 2.30 ± 0.46; cats without ATE = 2.04 ± 0.46. The bottom and top of the box are the first and third quartiles; the band within the box is the median. The whiskers correspond to the lowest datum still within 1.5 interquartile ranges (IQRs) of the first quartile, and the hightest datum still within 1.5 IQRs of the fourth quartile. Circles are outlier values (>1.5 IQRs away from the closest end of the box). Asterisks are extreme outlier values (>3 IQRs away from the closest end of the box)

Multivariable logistic regression model for the risk of ATE occurrence

The final multivariable model included the following variables with significant P values in univariate analysis: sex and neuter status (with male neutered cats as the reference category), presence and type of CHF (with cats with cardiac disease but no evidence of CHF as the reference category), type of cardiac disease (with HCM as the reference category) and LA:Ao. The multivariable logistic regression model (Logit ATE = −4.22 + 1.43 _* LA:Ao + 1.29 _*Group_G3vsG1+ 1.90 _* Group_G3vsG2) indicated that cats with cardiac disease but without effusion (ie, belonging to groups of cats with cardiac disease but no evidence of CHF and with evidence of cardiogenic pulmonary oedema) and an increase in LA:Ao were significantly associated with higher risk of developing ATE. Sex and neuter status and type of cardiac disease did not affect risk of ATE occurrence (Table 6).

Table 6

Result of multivariable logistic regression model for the risk of developing arterial thromboembolism

Variable	P value	OR (95% CI)
Groups
1 (without CHF) (n = 179)	0.005	3.64 (1.46–9.04)
2 (pulmonary edema) (n = 66)	<0.001	6.65 (2.59–17.06)
3 (pleural effusion) (n = 121)	– (reference category)	–
LA:Ao	<0.001	4.20* (2.05–8.59)
Type of cardiac disease	0.606	–
Sex and neuter status	0.767	–

Odds ratio for each 1.0 unit of LA:Ao increase above the threshold value of 1.5

CHF = congestive heart failure; CI = confidence interval; LA:Ao = left atrial to aortic root ratio; OR = odds ratio

Discussion

The aim of this cross-sectional study was to determine whether cats with evidence of cardiogenic pleural effusion had a lower risk of developing ATE compared with cats with evidence of cardiogenic pulmonary oedema or with cardiac disease but no evidence of CHF. The results of the multivariable analysis support this hypothesis, showing that the risk of ATE in cats with evidence of cardiogenic pleural effusion is significantly lower than that of cats with evidence of cardiogenic pulmonary oedema and of cats with cardiac disease but no evidence of CHF. Moreover, the lower prevalence of ATE in cats with evidence of cardiogenic pleural effusion in comparison with cats with cardiac disease but no evidence of CHF (6.6% and 15.6%, respectively) suggests that the reduced risk of ATE development in cats with evidence of cardiogenic pleural effusion is not offset by the increased risk of left atrium enlargement in these cats, which represents a known risk factor for ATE,^1,2,4,6 at least when only cats with LA:Ao ⩾1.5 are studied.

To our knowledge, this is the first study to have examined the relationship between the presence of pleural effusion and ATE in cats with cardiac disease. In previous studies of cats with ATE, echocardiography was not performed in all patients and ATEs of non-cardiogenic origin were also included.^5,6,33,34 In addition, the presence or absence of CHF and the type of CHF (ie, pulmonary oedema vs pleural effusion) were not reported in the cats that underwent cardiac investigations.^5,6,33,34

In both uni- and multivariable analysis, cats with evidence of cardiogenic pleural effusion showed a significantly lower risk of developing ATE compared with cats with evidence of cardiogenic pulmonary oedema and an even lower risk compared with cats with cardiac disease but no evidence of CHF. This may suggest that the presence of cardiogenic pleural effusion could be protective against ATE. Although our study was not designed to investigate the causal relationship between cardiogenic pleural effusion and ATE, there are two possible explanations for this finding: cats with cardiogenic pleural effusion may have enhanced systemic fibrinolysis or they may have less systemic inflammation. In heart failure, pleural effusions contain the proteins and enzymes that participate in coagulation and fibrinolysis¹² in an environment (ie, the pleural space) where their actions are no longer well-regulated, resulting in the formation of fluid with an inherently increased fibrinogenolytic/fibrinolytic activity.^26,35–41 Therefore, the reduced risk of ATE occurrence in cats with evidence of cardiogenic pleural effusion compared with cats with evidence of cardiogenic pulmonary oedema and those with cardiac disease but no evidence of CHF could reflect a reduced capacity for thrombi formation in these cats due to increased systemic fibrinolytic activity induced by the cardiogenic pleural effusion, upon re-entering the systemic circulation. An alternative hypothesis for the reduced prevalence of ATE in cats with evidence of cardiogenic pleural effusion is that these cats have less severe inflammation than those with cardiac disease but no evidence of CHF or with evidence of CHF and pulmonary oedema. It is known that ‘cross-talk’ between the inflammatory and haemostatic systems is responsible for the activation of haemostasis associated with systemic inflammation.^42,43 However, although there are studies showing a general increase in inflammation in cats with cardiac disease, especially when CHF is present,⁴⁴ no studies have specifically investigated whether the degree of inflammation differs between cats with cardiac disease but no evidence of CHF, those with evidence of CHF and pulmonary oedema, and those with evidence of CHF and pleural effusion. The potential roles of the enhanced systemic fibrinolysis and of inflammation in the pathogenesis of cardiogenic ATE formation should be investigated by the measurement of fibrinolytic markers and acute phase proteins in these cats in future studies.

As has been previously shown, we found that cats with evidence of CHF (with pulmonary oedema and/or pleural effusion) had a larger left atrium compared with cats with cardiac disease but no evidence of CHF,^45,46 and also that left atrial enlargement was a risk factor for feline ATE.^1,2,4,6 Blood stasis,¹¹ local accumulation of activated coagulation factors and platelets,⁴⁷ LA endothelial damage secondary to turbulent flow⁷ causing increase platelet adhesion⁴⁸ and exposure of subendothelial tissue factor (the initiator of the clotting cascade)⁴⁹ may all contribute to the association between LA enlargement and the increased risk of cardiogenic ATE formation.

For the same LA size, cats with cardiogenic pleural effusion have poorer LA function compared with those without effusion.⁵⁰ Therefore, it might be expected that cats with cardiogenic pleural effusions would have had more ATE events than those with evidence of pulmonary oedema and those with cardiac disease but no evidence of CHF; however, this was not what our study showed. It may be therefore that the risk of ATE associated with other echocardiographic LA functional parameters that have previously been shown to be predictive of ATE^4,11 differs depending on whether pleural effusion, pulmonary oedema or CHF is present.

ATE was detected at presentation in 9.5% of the cats eligible for the study and rose to 15.8% after removing cats that did not meet the inclusion criteria. The highest prevalence of ATE was found in cats with HCM (20.0%). Ferasin et al⁵¹ reported a similar prevalence of ATE (7.5%) to our eligible population, in a group of cats with various cardiomyopathies. Studies in cats with only HCM have reported an ATE prevalence in the range of 5–17%.^2,4,52 However, this study was not designed to provide information on why different cardiac diseases have a different prevalence of ATE. It is possible that some of these differences, present in univariate analysis, were due to the excessive fragmentation in multiple small categories (ie, very low counts) of the variable ‘type of cardiac diseases’. This would explain why this variable no longer influenced the risk of ATE when analysed by the more robust multivariable statistical model.

It is interesting to note that different types of cardiac disease have different severities and types of CHF; however, this study was not designed to explain this finding and the authors have no explanation for this result.

This study has several limitations. First, due to its retrospective nature, some cats with cardiac diseases were excluded because the medical records were incomplete. In addition, the type of CHF may have been miscategorised in a low number of cases as a small volume of pleural effusion may have been missed on thoracic radiographs, especially if concurrent pulmonary oedema was present. However, as nearly 50% of the cats classified with evidence of cardiogenic pleural effusion had the effusion confirmed by thoracocentesis, and in almost all there was echocardiographic proof of pleural effusion, it is unlikely that cats were misclassified into this group. Second, to increase specificity of cardiogenic ATE in the studied population, cats with LA:Ao <1.5 were excluded. This strategy limited the sample size of the study and eliminated the possibility of evaluating what happens when the LA:Ao is below the chosen threshold. Nevertheless, the replication of the study with all 617 eligible cats would have led to a very similar result in the multivariable logistic regression model (ie, Logit ATE = −5.46 + 1.84 _* LA:Ao + 1.26_* Group_G3vsG1 + 2.03_* Group_G3vsG2). Third, a very small number of cats (approximately 5% of the entire studied population) were on cardiac treatment at the time of first presentation and this factor was not analysed for ATE occurrence. A final limitation of this study is that the sex and neuter status, age and type of cardiac disease were not homogenously distributed among the three groups, and these factors may have also influenced the occurrence of ATE. However, of these variables, only the type of cardiac disease and sex and neuter status were associated with ATE occurrence in univariate analysis and these were no longer associated with this outcome when analysed by multivariable logistic regression.

Conclusions

Cats with ATE had a significantly larger left atrium compared with cats without ATE, and LA dimension represents a risk factor for this outcome. In addition, the presence of cardiogenic pleural effusion is associated with a lower risk of developing ATE in cats with cardiac disease. These results should be confirmed in future studies with prospective population enrolment.

Footnotes

Acknowledgements

The authors thank Dr Alix McBrearty for the language editing of this paper.

Author note

This paper was presented, in part, as an oral abstract at the European College of Veterinary Internal Medicine – Companion Animal Congress, 2019, Milan.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The work described in this manuscript involved the use of non-experimental (owned or unowned) animals. Established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient were always followed and/or this work involved the use of cadavers. Ethical approval from a committee was therefore not specifically required for publication in JFMS. Although not required, where ethical approval was still obtained, it is stated in the manuscript.

Informed consent

Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers) for all procedure(s) undertaken (prospective or retrospective studies). No animals or people are identifiable within this publication, and therefore additional informed consent for publication was not required.

ORCID iD

Andrea Zoia

References

Laste

Harpster

. A retrospective study of 100 cases of feline distal aortic thromboembolism: 1977–1993. J Am Anim Hosp Assoc 1995; 31: 492–500.

Rush

Freeman

Fenollosa

, et al. Population and survival characteristics of cats with hypertrophic cardiomyopathy: 260 cases (1990–1999). J Am Vet Med Assoc 2002; 220: 202–207.

Smith

Tobias

. Feline arterial thromboembolism: an update. Vet Clin North Am Small Anim Pract 2004; 34: 1245–1271.

Payne

Borgeat

Brodbelt

, et al. Risk factors associated with sudden death vs. congestive heart failure or arterial thromboembolism in cats with hypertrophic cardiomyopathy. J Vet Cardiol 2015; 17 Suppl 1: S318–328.

Fox

Keene

Lamb

, et al. International collaborative study to assess cardiovascular risk and evaluate long-term health in cats with preclinical hypertrophic cardiomyopathy and apparently healthy cats: the REVEAL study. J Vet Intern Med 2018; 32: 930–943.

Schoeman

. Feline distal aortic thromboembolism: a review of 44 cases (1990–1998). J Feline Med Surg 1999; 1: 221–231.

Liu

. Acquired cardiac lesions leading to congestive heart failure in the cat. Am J Vet Res 1970; 31: 2071–2088.

Helenski

Ross

. Platelet aggregation in feline cardiomyopathy. J Vet Intern Med 1987; 1: 24–28.

Bédard

Lanevschi-Pietersma

Dunn

. Evaluation of coagulation markers in the plasma of healthy cats and cats with asymptomatic hypertrophic cardiomyopathy. Vet Clin Pathol 2007; 36: 167–172.

10.

Stokol

Brooks

Rush

, et al. Hypercoagulability in cats with cardiomyopathy. J Vet Intern Med 2008; 22: 546–552.

11.

Schober

Maerz

. Assessment of left atrial appendage flow velocity and its relation to spontaneous echocardiographic contrast in 89 cats with myocardial disease. J Vet Intern Med 2006; 20: 120–130.

12.

Glauser

Otis

Levine

, et al. Coagulation factors and fibrinogen in pleural effusions. Respiration 1976; 33: 396–402.

13.

Henderson

Stein

Kutner

, et al. Analysis of twenty-three plasma proteins in ascites. The depletion of fibrinogen and plasminogen. Ann Surg 1980; 192: 738–742.

14.

Zoia

Drigo

Piek

, et al. Hemostatic findings in ascitic fluid: a cross-sectional study in 70 dogs. J Vet Intern Med 2017; 31: 43–50.

15.

Agarwal

Joyner

Swaim

. Ascites fluid as a possible origin for hyperfibrinolysis in advanced liver disease. Am J Gastroenterol 2000; 95: 3218–3224.

16.

Diab

Fathy

Soliman

, et al. Hyperfibrinolysis in advanced liver disease: does ascites have a role? Arab J Gastroenterol 2005; 6: 124–130.

17.

Delgado

Monreal

Armengou

, et al. Peritoneal D-dimer concentration for assessing peritoneal fibrinolytic activity in horses with colic. J Vet Intern Med 2009; 23: 882–889.

18.

Idell

Zwieb

Kumar

, et al. Pathways of fibrin turnover of human pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol 1992; 7: 414–426.

19.

Ivarsson

Holmdahl

Falk

, et al. Characterization and fibrinolytic properties of mesothelial cells isolated from peritoneal lavage. Scand J Clin Lab Invest 1998; 58: 195–204.

20.

Iakhiaev

Idell

. Activation and degradation of protein C by primary rabbit pleural mesothelial cells. Lung 2006; 184: 81–88.

21.

Mutsaers

Wilkosz

. Structure and function of mesothelial cells. In: Ceelen

(ed). Peritoneal carcinomatosis: a multidisciplinary approach. Boston, MA: Springer US, 2007, pp 1–19.

22.

Zoia

Drigo

Piek

, et al. Hemostatic findings of pleural fluid in dogs and the association between pleural effusions and primary hyperfibrino(geno)lysis: a cohort study of 99 dogs. PLoS One 2018; 13: e0192371. DOI: 10.1371/journal.pone.0192371.

23.

Zoia

Drigo

Simioni

, et al. Association between ascites and primary hyperfibrinolysis: a cohort study in 210 dogs. Vet J 2017; 223: 12–20.

24.

Zoia

Drigo

Piek

, et al. Enhanced fibrinolysis detection in a natural occurring canine model with intracavitary effusions: comparison and degree of agreement between thromboelastometry and FDPs, D-dimer and fibrinogen concentrations. PLoS One 2019; 14: e0225089. DOI: 10.1371/journal.pone.0225089.

25.

Zoia

Augusto

Drigo

, et al. Evaluation of hemostatic and fibrinolytic markers in dogs with ascites attributable to right-sided congestive heart failure. J Am Vet Med Assoc 2012; 241: 1336–1343.

26.

Chapin

Hajjar

. Fibrinolysis and the control of blood coagulation. Blood Rev 2015; 29: 17–24.

27.

Ferasin

DeFrancesco

. Management of acute heart failure in cats. J Vet Cardiol 2015; 17 Suppl 1: S173–189.

28.

Quinn

Sheley

Semonsen

, et al. Aortic and lower-extremity arterial disease: evaluation with MR angiography versus conventional angiography. Radiology 1998; 206: 693–701.

29.

Klainbart

Kelmer

Vidmayer

, et al. Peripheral and central venous blood glucose concentrations in dogs and cats with acute arterial thromboembolism. J Vet Intern Med 2014; 28: 1513–1519.

30.

Abbott

MacLean

. Two-dimensional echocardiographic assessment of the feline left atrium. J Vet Intern Med 2006; 20: 111–119.

31.

Zoia

Slater

Heller

. A new approach to pleural effusion in cats: markers for distinguishing transudates from exudates. J Feline Med Surg 2009; 11: 847–855.

32.

Zoia

Drigo

. Diagnostic value of Light’s criteria and albumin gradient in classifying the pathophysiology of pleural effusion formation in cats. J Feline Med Surg 2016; 18: 666–672.

33.

Smith

Dukes-McEwan

. Clinical signs and left atrial size in cats with cardiovascular disease in general practice. J Small Anim Pract 2012; 53: 27–33.

34.

Borgeat

Wright

Garrod

, et al. Arterial thromboembolism in 250 cats in general practice: 2004–2012. J Vet Intern Med 2014; 28: 102–108.

35.

Denny

Minot

. The coagulation of blood in the pleural cavity. Am J Physiol Content 1916; 39: 455–458.

36.

Napoli

Symbas

Vroon

, et al. Autotransfusion from experimental hemothorax: levels of coagulation factors. J Trauma 1987; 27: 296–300.

37.

Boadie

Glover

Bang

, et al. Clotting competence of intracavitary blood in trauma victims. Ann Emerg Med 1981; 10: 127–130.

38.

Schved

Gris

Gilly

, et al. Etude de l’activité fibrinolytique dans les liquides de ponction d’hémothorax traumatique [article in French]. Ann Fr Anesth Reanim 1991; 10: 104–107.

39.

Idell

Girard

Koenig

, et al. Abnormalities of pathways of fibrin turnover in the human pleural space. Am Rev Respir Dis 1991; 144: 187–194.

40.

Philip-Joet

Alessi

Philip-Joet

, et al. Fibrinolytic and inflammatory processes in pleural effusions. Eur Respir J 1995; 8: 1352–1356.

41.

Mutsaers

Birnie

Lansley

, et al. Mesothelial cells in tissue repair and fibrosis. Front Pharmacol 2015; 6. DOI: 10.3389/fphar.2015.00113.

42.

Levi

Keller

Van Gorp

, et al. Infection and inflammation and the coagulation system. Cardiovasc Res 2003; 60: 26–39.

43.

Smith

. Overview of hemostasis. In: Weiss

Wardrop

(eds). Schalm’s veterinary hematology. 6th ed. Ames, IA: Wiley-Blackwell, 2010, pp 635–653.

44.

Liu

Köster

Fosgate

, et al. Cardiovascular-renal axis disorder and acute-phase proteins in cats with congestive heart failure caused by primary cardiomyopathy. J Vet Intern Med 2020; 34: 1078–1090.

45.

Duler

Scollan

LeBlanc

. Left atrial size and volume in cats with primary cardiomyopathy with and without congestive heart failure. J Vet Cardiol 2019; 24: 36–47.

46.

Linney

Dukes-Mcewan

Stephenson

, et al. Left atrial size, atrial function and left ventricular diastolic function in cats with hypertrophic cardiomyopathy. J Small Anim Pract 2014; 55: 198–206.

47.

Hathcock

. Flow effects on coagulation and thrombosis. Arterioscler Thromb Vasc Biol 2006; 26: 1729–1737.

48.

Fox

Sisson

Moıse

. Textbook of canine and feline cardiology: principles and practice. 2nd ed. Philadelphia, PA: WB Saunders, 1999, pp 621–678.

49.

Eilertsen

K-E

Østerud

. Tissue factor. Blood Coagul Fibrinolysis 2004; 15: 521–538.

50.

Johns

Nelson

Gay

. Left atrial function in cats with left-sided cardiac disease and pleural effusion or pulmonary oedema. J Vet Intern Med 2012; 26: 1134–1139.

51.

Ferasin

Sturgess

Cannon

, et al. Feline idiopathic cardiomyopathy: a retrospective study of 106 cats (1994–2001). J Feline Med Surg 2003; 5: 151–159.

52.

Atkins

Gallo

Kurzman

, et al. Risk factors, clinical signs, and survival in cats with a clinical diagnosis of idiopathic hypertrophic cardiomyopathy: 74 cases (1985–1989). J Am Vet Med Assoc 1992; 201: 613–618.