Murmur intensity in adult dogs with pulmonic and subaortic stenosis reflects disease severity

A machine-learning algorithm to grade heart murmurs and stage preclinical myxomatous mitral valve disease in dogs

BACKGROUND

The presence and intensity of heart murmurs are sensitive indicators of several cardiac diseases in dogs, particularly myxomatous mitral valve disease (MMVD), but accurate interpretation requires substantial clinical expertise.

OBJECTIVES

Assess if a machine-learning algorithm can be trained to accurately detect and grade heart murmurs in dogs and detect cardiac disease in electronic stethoscope recordings.

ANIMALS

Dogs (n = 756) with and without cardiac disease attending referral centers in the United Kingdom.

METHODS

All dogs received full physical and echocardiographic examinations by a cardiologist to grade any murmurs and identify cardiac disease. A recurrent neural network algorithm, originally trained for heart murmur detection in humans, was fine-tuned on a subset of the dog data to predict the cardiologist's murmur grade from the audio recordings.

RESULTS

The algorithm detected murmurs of any grade with a sensitivity of 87.9% (95% confidence interval [CI], 83.8%-92.1%) and a specificity of 81.7% (95% CI, 72.8%-89.0%). The predicted grade exactly matched the cardiologist's grade in 57.0% of recordings (95% CI, 52.8%-61.0%). The algorithm's prediction of loud or thrilling murmurs effectively differentiated between stage B1 and B2 preclinical MMVD (area under the curve [AUC], 0.861; 95% CI, 0.791-0.922), with a sensitivity of 81.4% (95% CI, 68.3%-93.3%) and a specificity of 73.9% (95% CI, 61.5%-84.9%).

CONCLUSION AND CLINICAL IMPORTANCE

A machine-learning algorithm trained on humans can be successfully adapted to grade heart murmurs in dogs caused by common cardiac diseases, and assist in differentiating preclinical MMVD. The model is a promising tool to enable accurate, low-cost screening in primary care.

Self-Reported Management of Incidentally Detected Heart Murmurs in Puppies: A Survey among Veterinarians

BACKGROUND

Heart murmurs in puppies can be innocent or pathologic; the latter is almost always related to a congenital heart disease. Differentiating between these murmurs can be challenging for practicing veterinarians, but this differentiation is essential to ensure the best prognosis for puppies having a congenital heart disease. Our study aimed to reveal how veterinarians manage puppies with a heart murmur.

METHODS

A web-based questionnaire was sent to Dutch and Belgian veterinary practices.

RESULTS

Data from 452 respondents were analyzed. Though 88% of the respondents find detecting a heart murmur easy, only 9% find differentiating innocent murmurs from pathologic murmurs in puppies easy. Of the respondents, only 80% recommend immediate additional examination when detecting a loud heart murmur during the first veterinary health check at 6 weeks of age. Most of the respondents are aware that normal growth and the absence of clinical signs do not exclude severe congenital heart disease. Of the respondents, 31% were uncertain whether early surgical intervention could lead to improved outcomes.

CONCLUSIONS

Veterinarians are aware of the importance of echocardiography for puppies with a loud heart murmur, and recognize their limitations when differentiating an innocent from a pathological heart murmur in a puppy.

Doppler Ultrasonographic Assessment of Abdominal Aortic Flow to Evaluate the Hemodynamic Relevance of Left-to-Right Shunting Patent Ductus Arteriosus in Dogs

In this multicenter, prospective, observational study, abdominal aortic flow was examined with pulsed-wave Doppler ultrasound in dogs with a left-to-right shunting patent ductus arteriosus (PDA) and in apparently healthy dogs. Forty-eight dogs with a PDA and 35 controls were included. In the dogs with a PDA, 37/48 had hemodynamically significant PDAs (hsPDAs) while 11/48 had non-hsPDAs, based on the presence or absence of echocardiographic signs of left-sided volume overload, respectively. In 12 dogs (4/35 control dogs, 7/37 dogs with an hsPDA and 1/11 dogs with a non-hsPDA), the diastole was too short to visualize the end-diastolic flow. Antegrade end-diastolic flow was observed in 30/35 controls and 6/11 dogs with a non-hsPDA. Absent end-diastolic flow was observed in 1/35 control dogs and 3/11 dogs with a non-hsPDA. Retrograde end-diastolic flow was observed in 30/37 dogs with an hsPDA and 1/11 dogs with a non-hsPDA. Twenty-one dogs (15 with an hsPDA and 6 with a non-hsPDA) were reassessed after PDA closure, and, in 19/21, end-diastolic flow was visualized: 17/19 showed an antegrade flow, 1/19 an absent flow and 1/19 a retrograde flow. Sensitivity and specificity of retrograde end-diastolic flow for detection of hsPDAs were 100% and 90%, respectively. In conclusion, ultrasonographic assessment of abdominal aortic flow was feasible in dogs with PDA. However, end-diastolic flow was not always visualized. The presence of a retrograde end-diastolic flow was an accurate finding for discriminating hsPDAs and non-hsPDAs.

Beyond Heart Murmur Detection: Automatic Murmur Grading From Phonocardiogram

OBJECTIVE

Murmurs are abnormal heart sounds, identified by experts through cardiac auscultation. The murmur grade, a quantitative measure of the murmur intensity, is strongly correlated with the patient's clinical condition. This work aims to estimate each patient's murmur grade (i.e., absent, soft, loud) from multiple auscultation location phonocardiograms (PCGs) of a large population of pediatric patients from a low-resource rural area.

METHODS

The Mel spectrogram representation of each PCG recording is given to an ensemble of 15 convolutional residual neural networks with channel-wise attention mechanisms to classify each PCG recording. The final murmur grade for each patient is derived based on the proposed decision rule and considering all estimated labels for available recordings. The proposed method is cross-validated on a dataset consisting of 3456 PCG recordings from 1007 patients using a stratified ten-fold cross-validation. Additionally, the method was tested on a hidden test set comprised of 1538 PCG recordings from 442 patients.

RESULTS

The overall cross-validation performances for patient-level murmur gradings are 86.3% and 81.6% in terms of the unweighted average of sensitivities and F1-scores, respectively. The sensitivities (and F1-scores) for absent, soft, and loud murmurs are 90.7% (93.6%), 75.8% (66.8%), and 92.3% (84.2%), respectively. On the test set, the algorithm achieves an unweighted average of sensitivities of 80.4% and an F1-score of 75.8%.

CONCLUSIONS

This study provides a potential approach for algorithmic pre-screening in low-resource settings with relatively high expert screening costs.

SIGNIFICANCE

The proposed method represents a significant step beyond detection of murmurs, providing characterization of intensity, which may provide an enhanced classification of clinical outcomes.

Hemodynamic Effects of Subaortic Stenosis on Blood Flow Characteristics of a Mechanical Heart Valve Based on OpenFOAM Simulation

Subaortic stenosis (SAS) is a common congenital heart disease that can cause significant morbidity and mortality if not treated promptly. Patients with heart valve disease are prone to complications after replacement surgery, and the existence of SAS can accelerates disease progression, so timely diagnosis and treatment are required. However, the effects of subaortic stenosis on mechanical heart valves (MHV) are unknown. This study aimed to investigate flow characteristics in the presence of subaortic stenosis and computationally quantify the effects on the hemodynamics of MHV. Through the numerical simulation method, the flow characteristics and related parameters in the presence of SAS can be more intuitively observed. Based on its structure, there are three types of SAS: Tunnel-type SAS (TSS); Fibromuscular annulus SAS (FSS); Discrete SAS (DSS). The first numerical simulation study on different types of SAS found that there are obvious differences among them. Among them, the tunnel-type SAS formed a separated vortex structure on the tunnel-type narrow surface, which exhibits higher wall shear force at a low obstacle percentage. However, discrete SAS showed obvious differences when there was a high percentage of obstacles, forming high peak flow, high wall shear stress, and a high-intensity complex vortex. The presence of all three types of SAS results in the formation of high-velocity jets and complex vortices in front of the MHV, leading to increased shear stress and stagnation time. These hemodynamic changes significantly increase the risk of MHV dysfunction and the development of complications. Despite differences between the three types of SAS, the resultant effects on MHV hemodynamics are consistent. Therefore, early surgical intervention is warranted in SAS patients with implanted MHV.

Clinical and Electrocardiographic Findings for Predicting the Severity of Pulmonary Valve Stenosis in Dogs

Pulmonary valve stenosis (PS) in dogs is usually suspected due to the presence of a heart murmur and clinical signs. Echocardiography is needed to confirm the diagnosis and define the severity of PS. This retrospective study evaluated the utility of clinical and electrocardiographic (ECG) findings in the prediction of PS severity. Data regarding heart murmur and ECG analysis were gathered. Ninety-seven dogs with PS were included. A murmur grade ≥IV/VI was predictive of severe PS (area under curve (AUC) = 0.71; sensitivity (Se) = 95%; specificity (Sp) = 33%; p = 0.003). In lead II, P wave amplitude >0.35 mV (AUC = 0.67; Se = 31%; Sp = 100%; p = 0.038), Q wave < 0.15 mV (AUC = 0.70; Se = 70%; Sp = 59%; p = 0.0015), R wave < 0.87 mV (AUC = 0.66; Se = 67%; Sp = 69%; p = 0.006), and S wave > 0.37 mV (AUC = 0.80; Se = 72%; Sp = 85%; p < 0.0001) were predictive of severe PS. The extent of right deviation of the mean electrical axis of the QRS complex was correlated with the pulmonary pressure gradient (r = 0.648; p < 0.0001). In conclusion, a systolic murmur with intensity ≥IV/VI, a P wave amplitude >0.35 mV, low amplitude of Q and R waves, deep S waves in lead II, and right axis deviation of the QRS complex in a young dog are predictive of severe PS.

Right axis deviation in the canine electrocardiogram for predicting severity of pulmonic stenosis: a retrospective cohort analysis

OBJECTIVE

To investigate the predictive value of right axis deviation of the mean electrical axis (MEA) in assessing the severity of pulmonic stenosis (PS) in dogs.

ANIMALS

Records for 218 client-owned dogs diagnosed between 2014 and 2020 with PS as determined by Doppler echocardiography.

PROCEDURES

University of Florida Small Animal Clinic medical records were reviewed, and signalment and clinical risk variables (murmur grade and clinical signs) were extracted. MEA was determined from ECG records by use of leads I and III. Predictive potential of MEA and associated risk factors to diagnose PS severity (mild [< 50 mm Hg], moderate, or severe [> 75 mm Hg]) were assessed by receiver-operating characteristic curve analysis and quantile regression.

RESULTS

Records for 88 dogs were eligible for analysis. Greater PS severity was associated with smaller breeds presenting with ECG abnormalities, overt clinical signs, and high-category murmur grades (IV and V). Mean MEA increased with stenosis severity category, with an average of 62° for mild, 113° for moderate, and 157° for severe. Each 10° increase in MEA corresponded to an approximately 5-mm Hg increase in PG. Increasing PS severity was associated with MEA right axis deviation > 100° and the more severe cases (PG > 75 mm Hg) with MEA right axis deviation > -180°.

CLINICAL RELEVANCE

Mean electrical axis right axis deviation may be a useful screening metric for dogs with suspected moderate to severe PS.

Association between Cardiac Auscultation and Echocardiographic Findings in Warmblood Horses

Heart murmurs are detected frequently when auscultating horses and certain murmurs can usually be linked to specific valvular regurgitations. Limited information exists about the accuracy of these broad rules in warmblood horses and the influence of grade of the regurgitation and dimensional changes on murmur intensity. This study aims to clarify the accuracy of cardiac auscultation in warmblood horses and the influence of the grade of regurgitation and dimensional changes on the loudness of the murmur. In this retrospective study, 822 warmblood horses presented for cardiac examination in a large equine referral center in northern Germany underwent a thorough cardiac auscultation. In total, 653 of these revealed one or more heart murmurs. Most common auscultatory findings were left-sided systolic murmurs (68%) or left-sided diastolic murmurs (15%). On 635 of these horses, an echocardiographic examination was performed, revealing regurgitations of the mitral valve as the most common valvular regurgitation (77%) followed by regurgitations of the aortic valve (23%). Thirty-one percent of horses that underwent echocardiographic examination displayed dimensional changes of one or more compartments of the heart, with the left atrium being most affected (21%), followed by the left ventricle (13%). The main goal of this study was to link certain auscultatory findings with results of the echocardiographic examinations, trying to determine whether auscultation and echocardiography agreed on the valve affected, as well as to find out if loudness of the murmur coincided with grade of regurgitation and presence of dimensional changes. Agreement between auscultation and cardiac ultrasound was substantial (Kappa 0.74) if one or more murmurs and regurgitations were present and almost perfect (Kappa 0.94) if only one murmur and one regurgitation were found. Auscultation was particularly well suited for detection of left-sided systolic and diastolic murmurs, with 87% of left-sided systolic murmurs being caused by a mitral valve regurgitation and 81% of left-sided diastolic murmurs originating from an aortic valve regurgitation. We found a fair agreement between the grade of regurgitation and the respective murmur. Association was particularly good between mild regurgitations and low-grade murmurs, while differentiation between moderate to severe regurgitation based upon the loudness of the murmur was less reliable. Dimensional changes were usually linked to more severe regurgitations and higher-grade murmurs. However, a direct correlation between murmur intensity and the presence or severity of dimensional changes, independent of the grade of valvular regurgitation, could not be established in this cohort of horses.

Natural history of subaortic stenosis in 166 dogs (1999-2011)

INTRODUCTION

Subaortic stenosis (SAS) is one of the most common congenital cardiac diseases in dogs. The objective of this study was to provide survival times on a large population of dogs with SAS and to propose a redefined pressure gradient (PG) scale to include a mild, moderate, severe and very severe disease group.

ANIMALS, MATERIALS AND METHODS

Dogs were divided into four groups based on the Doppler-derived PG across the stenosis. Disease severity was defined as follows: mild = PG < 50 mmHg; moderate = PG range 50-80 mmHg; severe = PG range 80-130 mmHg; and very severe = PG > 130 mmHg. Over the study period (1999-2011), 166 client-owned dogs were diagnosed with SAS of which 129 had follow-up information available.

RESULTS

Unadjusted median survival time for severity groups were as follows: mild 10.6 years; moderate 9.9 years; severe 7.3 years; and very severe 3.0 years. Univariable analysis examining the effect of the PG, age at diagnosis and sex found only the PG and age at diagnosis had a significant effect on survival. Adjusted survival curves showed that the survival time in the very severe group was decreased compared with all other groups.

CONCLUSION

Based on the results of this study, a revised SAS classification system with four PG groups is appropriate. Dogs with a PG > 130 mmHg were identified as those with the lowest median survival time.

Identification and Clinical Significance of Heart Murmurs in Puppies Involved in Puppy Trade

The detection of a congenital heart defect at purchase is an important step in early detection from a clinical and legal standpoint. Indeed, some cardiac abnormalities may be corrected with surgery, and very often, treatment needs to be performed early before congestive heart failure or irreversible heart damage can occur. From a legal viewpoint, if the defect is revealed in a newly purchased puppy, the buyer may be required to return it and receive compensation. Puppies affected with congenital heart defects are likely to die prematurely, causing emotional suffering to the owner. Furthermore, by considering breed predisposition, early recognition allows breeders to avoid breeding from particular dogs with genetic defects and prevent the continuation of genetic defects in breeding lines. Given gaps in the literature about the recognition of murmurs in the puppy trade, the present article describes how to identify a heart murmur in a puppy during a pre-purchase examination and its significance from a clinical and legal viewpoint. In the canine population, the prevalence of cardiac defects ranges between 0.13 and 1.6%. Pulmonic stenosis is the most common defect found in puppies, followed by patent ductus arteriosus, subaortic stenosis, and ventricular septal defect. On the basis of the above considerations, the veterinarian should recognize and identify the murmur following a protocol for routine examination of puppies involved in trade.

Digital phonocardiography of cardiac arrhythmias in dogs - Preliminary experiences

Electronic stethoscopes and digital phonocardiograms (DPCGs) can be applied when diagnosing cardiac murmurs, but their use for cardiac arrhythmias is not described in veterinary medicine. Data of 10 dogs are presented in this preliminary study, demonstrating the applicability of these techniques. Although the number of artefacts and the amount of baseline noise produced by the two digitising systems used did not differ, the Welch Allyn Meditron system or similar ones capable of simultaneous recording of electrocardiograms (ECGs) and DPCGs provide a better option for clinical research and education, whilst the 3M Littmann 3200 system might be more suitable for everyday clinical settings. A combined system with simultaneous phonocardiogram and ECG, especially with wireless transmission, might be a solution in the future.

Cardiac auscultation skills in final year veterinary students and recent veterinary graduates, referral hospital veterinary surgeons and veterinary cardiologists or cardiology residents

BACKGROUND

Cardiac auscultation is an important part of the physical examination. This study evaluated cardiac auscultation skills in veterinary students and compared their abilities to recent veterinary graduates, referral hospital veterinary surgeons and veterinary cardiologists or cardiology residents. In addition it compared their self-predicted quiz scores to their actual scores, evaluating if they could accurately predict their own performance level.

METHODS

A digital recording device was used to record auscultation sounds from 12 different patients with a diagnosis confirmed by a board-certified veterinary cardiologist. The sound files and associated phonocardiograms were uploaded to a video sharing website. A cloud-based online multiple-choice quiz was generated and shared with final year veterinary students, recent veterinary graduates, referral hospital veterinary surgeons and veterinary cardiologists or cardiology residents.

RESULTS

There were 128 participants: 51 final year veterinary students, 62 recent veterinary graduates, and 10 referral hospital veterinary surgeons and five veterinary cardiologists or cardiology residents. No difference was found between the cardiac auscultation skills of recent veterinary graduates and final year veterinary students. Veterinary students' self-predicted scores were lower than actual scores.

CONCLUSIONS

Recent veterinary graduates did not perform better than final year veterinary students in this study, suggesting that auscultation skills do not continue to improve in the first few years after graduation. Efforts should be made to maximise students' learning in cardiac auscultation skills. Veterinary students show a lack of confidence in cardiac auscultation skills.

Cardiomyopathy Associated With Coronary Arteriosclerosis in Free-Ranging Eurasian Lynx (Lynx lynx carpathicus)

The Eurasian lynx (subspecies Lynx lynx carpathicus) was reintroduced to Switzerland in the 1970's. Health monitoring of the reintroduced population started in the late 1980's. Since then, six lynx have been found affected by a myocardial disease. The earliest case was an animal that died after a field anesthesia. Two lynx were found dead, two were euthanized/culled because of disease signs, and one was hit by car. Two had a heart murmur at clinical examination. At necropsy, the first animal showed only lung edema but the other five had cardiomegaly associated with myocardial fibrosis. Three had multisystemic effusions. Histological examination of all six lynx showed mild to severe, multifocal, myocardial interstitial and perivascular fibrosis along with multifocal myocyte degeneration and loss, and replacement fibrosis. Moderate to severe multifocal arteriosclerosis with associated luminal stenosis of the small and medium-sized intramural coronary arteries and the presence of Anitschkow cells was also observed. The heart lesions may have led to sudden death in the first case and to a chronic right-sided heart failure in the remaining. None of the lynx showed lesions or signs suggestive of an acute or subacute infection. Given the common geographic origin of these animals and the severe loss of heterozygocity in this population, a genetic origin of the disease is hypothesized.

Epidemiological study of congenital heart diseases in dogs: Prevalence, popularity, and volatility throughout twenty years of clinical practice

The epidemiology of Congenital Heart Diseases (CHDs) has changed over the past twenty years. This study aimed to evaluate the prevalence of CHDs in the population of dogs recruited in a single referral center (RC); compare the epidemiological features of CHDs in screened breeds (Boxers) versus non-screened (French and English Bulldogs, German Shepherds); investigate the association of breeds with the prevalence of CHDs; determine the popularity and volatility of breeds over a 20-year period; analysed the trends of the most popular breeds in the overall population of new-born dogs registered in the Italian Kennel Club (IKC) from 1st January 1997 to 31st December 2017. The RC's cardiological database was analysed, and 1,779 clinical records were included in a retrospective observation study. Descriptive statistics and frequencies regarding the most representative breeds and CHDs were generated. A logistic regression model was used to analyse the trends of the most common CHDs found in single and in cluster of breeds. The relationship between breed popularity and presence of CHDs was studied. The most common CHDs were Pulmonic Stenosis, Patent Ductus Arteriosus, Subaortic Stenosis, Ventricular Septal Defect, Aortic Stenosis, Tricuspid Dysplasia, Atrial Septal Defect, Double Chamber Right Ventricle, Mitral Dysplasia, and others less frequent. The most represented pure breeds were Boxer, German Shepherd, French Bulldog, English Bulldog, Maltese, Newfoundland, Rottweiler, Golden Retriever, Chihuahua, and others in lower percentage. Chihuahuas, American Staffordshire Terriers, Border Collies, French Bulldogs, and Cavalier King Charles Spaniel were the most appreciated all of which showed a high value of volatility. This study found evidence for the value of the screening program implemented in Boxers; fashions and trends influence dog owners' choices more than the worries of health problems in a breed. Effective breeding programs are needed in order to control the diffusion of CHDs without impoverishing the genetic pool.

Breed does not affect the association between murmur intensity and disease severity in dogs with pulmonic or subaortic stenosis

OBJECTIVES

To determine whether breed affects the ability of murmur intensity to predict the severity of stenosis in dogs with pulmonic stenosis or subaortic stenosis.

MATERIALS AND METHODS

Retrospective multi-investigator study of dogs with pulmonic stenosis or subaortic stenosis. Murmur intensity, assessed by a four-level classification scheme, was compared with echocardiographically-determined pressure gradient across the affected valve. Breeds represented by at least 10 dogs at any murmur intensity were compared to determine the effect, if any, of breed.

RESULTS

A total of 1088 dogs (520 with pulmonic stenosis and 568 with subaortic stenosis, representing 106 breeds and the mixed breed group) were included; 208 dogs had soft, 210 had moderate, 283 had loud and 387 had palpable murmurs. Fifteen breeds were represented by at least 10 dogs: five breeds with at least 10 dogs had soft murmurs (132 dogs), nine breeds had moderate murmurs (149 dogs), 10 breeds had loud murmurs (188 dogs), and 11 breeds had palpable murmurs (286 dogs). No breeds differed in stenosis severity from any other breeds within any murmur grade. Post hoc power calculations suggested that we would have been able to detect at least a moderate or large effect size, had one existed. Several dogs with soft murmurs had more-than-mild disease severity.

CLINICAL SIGNIFICANCE

Despite anecdotally perceived differences in the detection of heart murmurs between breeds, which have been proposed to potentially affect the interpretation of stenosis severity, we found no obvious breed effect in the ability to predict severity of stenosis.