Automated echocardiographic left ventricular dimension assessment in dogs using artificial intelligence: Development and validation.

BACKGROUND: Artificial intelligence (AI) could improve accuracy and reproducibility of echocardiographic measurements in dogs. HYPOTHESIS: A neural network can be trained to measure echocardiographic left ventricular (LV) linear dimensions in dogs. ANIMALS: Training dataset: 1398 frames from 461 canine echocardiograms from a single specialist center. VALIDATION: 50 additional echocardiograms from the same center. METHODS: Training dataset: a right parasternal 4-chamber long axis frame from each study, labeled by 1 of 18 echocardiographers, marking anterior and posterior points of the septum and free wall. VALIDATION DATASET: End-diastolic and end-systolic frames from 50 studies, annotated twice (blindly) by 13 experts, producing 26 measurements of each site from each frame. The neural network also made these measurements. We quantified its accuracy as the deviation from the expert consensus, using the individual-expert deviation from consensus as context for acceptable variation. The deviation of the AI measurement away from the expert consensus was assessed on each individual frame and compared with the root-mean-square-variation of the individual expert opinions away from that consensus. RESULTS: For the septum in end-diastole, individual expert opinions deviated by 0.12 cm from the consensus, while the AI deviated by 0.11 cm (P = .61). For LVD, the corresponding values were 0.20 cm for experts and 0.13 cm for AI (P = .65); for the free wall, experts 0.20 cm, AI 0.13 cm (P < .01). In end-systole, there were no differences between individual expert and AI performances. CONCLUSIONS AND CLINICAL IMPORTANCE: An artificial intelligence network can be trained to adequately measure linear LV dimensions, with performance indistinguishable from that of experts.

Effects of pimobendan in cats with hypertrophic cardiomyopathy and recent congestive heart failure: Results of a prospective, double-blind, randomized, nonpivotal, exploratory field study.

BACKGROUND: The benefits of pimobendan in the treatment of congestive heart failure (CHF) in cats with hypertrophic cardiomyopathy (HCM) have not been evaluated prospectively. HYPOTHESIS/OBJECTIVES: To investigate the effects of pimobendan in cats with HCM and recent CHF and to identify possible endpoints for a pivotal study. We hypothesized that pimobendan would be well-tolerated and associated with improved outcome. ANIMALS: Eighty-three cats with HCM and recently controlled CHF: 30 with and 53 without left ventricular outflow tract obstruction. METHODS: Prospective randomized placebo-controlled double-blind multicenter nonpivotal field study. Cats received either pimobendan (0.30 mg/kg q12h, n = 43), placebo (n = 39), or no medication (n = 1) together with furosemide (<10 mg/kg/d) with or without clopidogrel. The primary endpoint was a successful outcome (ie, completing the 180-day study period without a dose escalation of furosemide). RESULTS: The proportion of cats in the full analysis set population with a successful outcome was not different between treatment groups (P = .75). For nonobstructive cats, the success rate was 32% in pimobendan-treated cats versus 18.2% in the placebo group (odds ratio [OR], 2.12; 95% confidence interval [CI], 0.54-8.34). For obstructive cats, the success rate was 28.6% and 60% in the pimobendan and placebo groups, respectively (OR, 0.27; 95% CI, 0.06-1.26). No difference was found between treatments for the secondary endpoints of time to furosemide dose escalation or death (P = .89). Results were similar in the per-protocol sets. Adverse events in both treatment groups were similar. CONCLUSIONS AND CLINICAL IMPORTANCE: In this study of cats with HCM and recent CHF, no benefit of pimobendan on 180-day outcome was identified.

Two-dimensional echocardiographic estimates of left atrial function in healthy dogs and dogs with myxomatous mitral valve disease.

OBJECTIVES: To provide reference intervals for 2-dimensional linear and area-based estimates of left atrial (LA) function in healthy dogs and to evaluate the ability of estimates of LA function to differentiate dogs with subclinical myxomatous mitral valve disease (MMVD) and similarly affected dogs with congestive heart failure (CHF). ANIMALS: Fifty-two healthy adult dogs, 88 dogs with MMVD of varying severity. METHODS: Linear and area measurements from 2-dimensional echocardiographs in both right parasternal long and short axis views optimized for the left atrium were used to derive estimates of LA active emptying fraction, passive emptying fraction, expansion index, and total fractional emptying. Differences for each estimate were compared between healthy and MMVD dogs (based on ACVIM classification), and between MMVD dogs with subclinical disease and CHF that had similar LA dimensions. Diagnostic utility at identifying CHF was examined for dogs with subclinical MMVD and CHF. Relationships with bodyweight were assessed. RESULTS: All estimates of LA function decreased with increasing ACVIM stage of mitral valve disease (p<0.05) and showed negative relationships with increasing LA size (all r2 values < 0.2), except for LA passive emptying fraction, which did not differ or correlate with LA size (p=0.4). However, no index of LA function identified CHF better than measurements of LA size. Total LA fractional emptying and expansion index showed modest negative correlations with bodyweight. CONCLUSIONS: Estimates of LA function worsen with worsening MMVD but fail to discriminate dogs with CHF from those with subclinical MMVD any better than simple estimates of LA size.

Common two-dimensional echocardiographic estimates of aortic linear dimensions are interchangeable.

OBJECTIVE: To compare two echocardiographic methods of measuring aortic diameter in short-axis projections. METHODS: Right-parasternal short-axis 2-dimensional projections of the left atrium and aorta were obtained from dogs and cats undergoing routine cardiac evaluation. Two investigators measured the aortic valve linear dimension using 2 methods: along the commissure between the non-coronary and right-coronary cusps and along the commissure between the non-coronary and left-coronary cusps. Inter-observer and intra-observer variability and agreement were assessed by comparing blinded measurements with each method by 4 trained observers on a standardized set of images. Measurements were compared for agreement using the limits of agreement analysis. Variability between observers was compared by examining residuals and intraclass correlation. RESULTS: 274 canine and 100 feline aortic valve images were measured in the first part of the study. One observer demonstrated slight proportional bias, while the other observer showed more variability (less agreement). When results were pooled for both investigators, no bias was identified, and 95% limits of agreement were ±10% of the mean measurement for both species. In the second part of the study, 106 images were measured. Intraobserver variability was <4% for all observers. Inter-observer agreement was very high. Individual bias was identified in some observers, but was considered clinically inconsequential. Normalized differences between the 2 methods of measurement were below ±15% of the measured value for all observers. CONCLUSIONS: Our results show sufficient agreement between two common methods used to measure aortic linear dimensions to suggest that these methods are interchangeable.

Immunohistochemical evidence for expression of fast-twitch type sarco(endo)plasmic reticulum Ca2+ ATPase (SERCA1) in German shepherd dogs with dilated cardiomyopathy myocardium.

OBJECTIVES: Dilated cardiomyopathy (DCM) is one of the most common acquired canine heart diseases. It is particularly common in large and giant breed dogs. Although a great deal is known about the clinical progression and manifestations of the disease, the underlying cellular and molecular mechanisms remain poorly understood. One widely held belief is that calcium-handling abnormalities are critically involved in the disease process. This study investigates the changes in expression of the sarco(endo)plasmic reticulum calcium ATPase (SERCA) isoforms in DCM myocardium from German shepherd dogs. ANIMALS, MATERIALS AND METHODS: Affected tissue samples were obtained from German shepherd dogs with DCM, euthanized for intractable congestive heart failure while normal myocardial tissue samples were obtained from German shepherd dogs, euthanized for non-cardiovascular reasons. Tissue microarrays containing normal and DCM myocardium samples were prepared, immunostained with SERCA1 and SERCA2 antibodies and analyzed. RESULTS: We were able to demonstrate, for the first time, that while there is little change in the expression of the cardiac isoform (SERCA2), there is clear expression of the fast-twitch skeletal muscle isoform SERCA1 in the myocardium of dogs diagnosed with DCM. CONCLUSION: We propose that SERCA1 expression is evidence of a natural adaptive response to the impaired Ca2+ handling thought to occur in German shepherd dogs with DCM and heart failure.

Counterpulsation from the skeletal muscle ventricle and the intraaortic balloon pump in the normal and failing circulations.

BACKGROUND: The intra-aortic balloon pump (IABP) is the device that is in most common use to provide cardiovascular support. A skeletal muscle ventricle (SMV) was configured to produce counterpulsation in the thoracic aorta similar to that obtained with an IABP. The hemodynamic effects of an IABP and a SMV in the same animal and in both normal and failing circulations were assessed. METHODS AND RESULTS: SMVs were connected to and IABPs were placed in the thoracic aorta of 12 anesthetized pigs. Hemodynamic parameters during the IABP- or the SMV-assisted beat were compared with those during the preassist beat. Acute heart failure was induced in 6 of the pigs by snaring the left anterior descending coronary artery (LAD). The hemodynamic effects of the IABP and the SMV were then reassessed. In the assisted cycles, SMV activation increased the mean aortic diastolic pressure (MADP) by 26.5+/-3.5%, the mean diastolic LAD flow by 48.4+/-7.2%, and endocardial viability ratio (EVR) by 31.6+/-3.8% (P<0.0001). In the same animals, IABP assist increased MADP by 19.8+/-2.3%, mean diastolic LAD flow by 37.2+/-3.9%, and EVR by 21.4+/-3.0% (P<0.0001). Under acute heart failure conditions, both SMV and IABP assist significantly increased MADP, mean diastolic LAD flow, and EVR. CONCLUSIONS: In both the normal and failing circulations, the SMV was an effective counterpulsator, providing cardiac assist that was at least equal to that available from an IABP. The SMV may therefore provide the proven benefits of an IABP in ambulant patients.

Pilot study on cardiac troponin I levels in dogs with pericardial effusion.

OBJECTIVES: The aim of this pilot study was to assess cardiac troponin I (cTnI) levels in pericardial effusion (PE) and plasma from dogs with PE. BACKGROUND: A reliable marker for detecting the etiology of PE in dogs remains undetermined. cTnI is becoming the gold standard marker for detecting myocardial damage in humans. ANIMALS, MATERIALS AND METHODS: Twenty-five dogs with PE (21 and 4 secondary to neoplasia and non-neoplasia causes, respectively) and 37 control dogs were studied. RESULTS: The median cTnI plasma level from 37 normal dogs versus 15 (out of 25) with PE was 0.03ng/mL and 0.19ng/mL, respectively (p<0.0001). The level of cTnI in PE versus plasma showed a significant correlation (p<0.01) with a Spearman r coefficient of 0.7603. No significant difference could be found upon comparison of dogs with only right atrial tumors (n=14) versus other types of neoplasia (n=7), nor between the group with right atrial tumors (n=14) versus all other cases including neoplasia as well as non-neoplasia (n=11). The median cTnI level in PE from dogs with neoplasia and non-neoplasia was not significantly different. CONCLUSIONS: cTnI did rise significantly in both PE and plasma in dogs with PE, but cTnI levels did not help differentiate between etiologies according to this study. One of the study groups is too small to allow final conclusions, and thus further investigation is warranted.

Aneurysmal dilatation of the right auricle in two dogs.

An 8-year-old sexually intact male Golden Retriever with a history of collapse during exercise underwent an examination during which tachydysrhythmia was identified. At another institution, a 12.5-year-old spayed female Lhasa Apso was referred because of a cough and for evaluation of a heart murmur. In the Golden Retriever, radiographic examination revealed bulging of the craniodorsal aspect of the cardiac silhouette and echocardiography revealed right atrial dilatation. In the Lhasa Apso, a cranial mediastinal mass was suspected on the basis of radiographic findings, but no abnormalities were detected echocardiographically. In both dogs, nonselective angiography and contrast-enhanced computed tomography revealed a markedly enlarged, thin-walled right auricle. Exploratory thoracotomy in the Golden Retriever revealed a defect in the pericardium through which the right auricle and part of the atrium had herniated. In dogs, a right auricular aneurysm should be considered in differential diagnoses of a cranial mediastinal mass (detected radiographically) adjacent to the cardiac silhouette.

Patent ductus arteriosus ligation and pulmonary artery banding in a kitten.

Patent ductus arteriosus (PDA) and ventricular septal defect were diagnosed in a 2-month-old, 0.9-kg, male kitten. This combination of PDA and ventricular septal defect in a cat has not been previously described. Surgical ligation of the PDA, together with palliation of the ventricular septal defect by pulmonary artery banding with a silastic band, successfully improved the kitten's condition.

Ladder diagrams for atrial flutter and atrial fibrillation.

Ladder diagrams are valuable teaching and interpretive tools in both human and veterinary electrocardiography. The purpose of this paper is to demonstrate how ladder diagrams can be used to help understand the possible mechanisms underlying atrial flutter and atrial fibrillation.

Echocardiography in the chinchilla.

Heart disease has been described in the chinchilla and, with increasing popularity as a pet, the demand for diagnostic evaluation and treatment has increased. The goal of this study was to determine reference values for echocardiographic measurements in chinchillas and the effect of anesthesia on these measurements. Seventeen clinically healthy adult chinchillas were studied. All animals were anesthetized with isoflurane by mask. Standard echocardiographic views were used. A difference was seen in the echocardiographic measurements for left ventricular systolic dimension, fractional shortening, aortic (Ao) diameter, left atrial (LA) diameter, ratio of LA diameter to Ao diameter, and peak flow velocities and ejection times for Ao and pulmonary artery flows between awake and anesthetized chinchillas.

Appendicular arterial tumor embolization in two cats with pulmonary carcinoma.

A 13-year-old neutered male Persian cat and an 11-year-old neutered female Persian cat were examined because of an acute onset of lameness. In both cats, conscious proprioception and reflexes were diminished in the affected limb. In 1 cat, no blood flow was detected in the left brachial artery with a Doppler ultrasonic flow detector, whereas blood flow in the right brachial artery was easily documented. In the other cat, the right femoral pulse was not palpable. Neither cat had any echocardiographic evidence of cardiac disease. In both cats, treatment was primarily supportive. One cat died, and the other was euthanatized. At necropsy, lung lobe consolidation was seen. Microscopically, there was multifocal infiltration of the lung parenchyma with cuboidal to columnar neoplastic epithelial cells. Neoplastic epithelial cells of similar morphology were identified in nodular masses in sections of muscle, and intravascular tumor emboli were identified obliterating small and large arterioles. Immunohistochemical staining of pulmonary and muscular tissue for pan-cytokeratin antigen revealed intense cytoplasmic staining of neoplastic cells. Staining for factor VIII-related antigen confirmed that clusters of neoplastic cells represented intravascular emboli. Clinical signs in the cats were attributed to arterial occlusion by tumor emboli.

Cardiac troponin I in feline hypertrophic cardiomyopathy.

Measurement of plasma cardiac troponin I concentration ([cTnI]) is a sensitive and specific means for detecting myocardial damage in many mammalian species. Studies have shown that [cTnI] increases rapidly after cardiomyocyte injury. The molecular structure of cTnl is highly conserved across species, and current assays developed for its detection in humans have been validated in many species. In this study, [cTnI] was quantified using a 2-site sandwich assay in plasma of healthy control cats (n = 33) and cats with moderate to severe hypertrophic cardiomyopathy (HCM) (n = 20). [cTnI] was significantly higher in cats with HCM (median, 0.66 ng/mL; range, 0.05-10.93 ng/mL) as compared with normal cats (median, <0.03 ng/mL; range, <0.03-0.16 ng/mL) (P < .0001). An increase in [cTnI] was also highly sensitive (sensitivity = 85%) and specific (specificity = 97%) for differentiating cats with moderate to severe HCM from normal cats. [cTnI] was weakly correlated with diastolic thickness of the left ventricular free wall (r2 = .354; P = .009) but not with the diastolic thickness of the interventricular septum (P = .8467) or the left atrium: aorta ratio (P = .0652). Furthermore, cats with congestive heart failure at the time of cTnI analysis had a significantly higher [cTnI] than did cats that had never had heart failure and those whose heart failure was controlled at the time of analysis (P = .0095 and P = .0201, respectively). These data indicate that cats with HCM have ongoing myocardial damage. Although the origin of this damage is unknown, it most likely explains the replacement fibrosis that is consistently identified in cats with moderate to severe HCM.