The reported prevalence of 0.1–0.5% for congenital cardiac defects in horses is relatively low as compared with other domestic animals.1–3 Cardiac defects account for approximately 3.5% of all congenital defects in horses.4 Such defects could have a profound impact on foals through altered growth rate, exercise intolerance, respiratory distress, and death. As such, it is important to recognize the clinical findings associated with such defects early in the course of medical evaluation of neonatal foals, as the presence of a defect can alter the prognosis and course of treatment. The available reports in the peer-reviewed literature include pathology reports2–5 single case reports, and small case series describing congenital cardiac defects in ≤5 foals.1,6–33 Authors of case reports have suggested an association between the Arabian breed and the occurrence of cardiac defects, but this suspicion has not been confirmed.1,22–34 There is a paucity of information available about the most prevalent cardiac defects, concurrent noncardiac congenital anomalies, and clinical findings in foals with cardiac anomalies. The purpose of this retrospective study was to identify the cardiac defects in a group of foals examined over a 15-year time period (1992–2007) and to review the signalment, presenting complaint, physical examination findings, diagnostic evaluation, concurrent noncardiac congenital anomalies, and outcome in affected foals.
Background: Literature available regarding congenital cardiac defects in foals is limited to reports of individual cases or small case series.
Objective: To describe the clinical, echocardiographic, and necropsy findings and breed predilection of congenital cardiac defects in neonatal foals.
Animals: Eighteen foals <15 days of age with 1 or more congenital cardiac defects.
Methods: Medical records of foals diagnosed with congenital cardiac defects at the William R. Pritchard Veterinary Medical Teaching Hospital were reviewed. Data collected included history, signalment, clinical signs, laboratory data, diagnostic and necropsy results, and outcome.
Results: Arabian foals represented 39% of cases with congenital cardiac defects and were significantly (P= .004) overrepresented (OR = 4.7 [CI: 1.8–12.4]) compared with the general hospital population. Ventricular septal defect (VSD) (14/18), tetralogy of Fallot (5/18), and tricuspid valve atresia (4/18) were the most common defects identified. A ≥3/6 heart murmur (14/14) accompanied by tachycardia (14/17), tachypnea (17/17), and cyanosis of mucous membranes (7/16) were the most common clinical signs. Concurrent congenital defects were common (9/18). Two foals, both with VSD, survived for ≥ 8 years after diagnosis and 1 was a successful performance horse.
Conclusions and Clinical Relevance: Arabian horses appear to have a predisposition for cardiac defects. The presence of a loud murmur (≥ 3/6), cyanotic membranes, and tachycardia or tachypnea in a neonatal foal should warrant thorough evaluation of the heart for congenital defects. Foals with cardiac defects should be closely evaluated for concurrent congenital defects in other body systems.
atrial septal defect
left heart base
mitral valve dysplasia
patent ductus arteriosus
patent foramen ovale
pulmonary valve atresia
total anomalous pulmonary venous return
tetralogy of Fallot
tricuspid valve atresia
ventricular septal defect
Materials and Methods
Medical records of foals admitted to the William R. Pritchard Veterinary Medical Teaching Hospital at the University of California, Davis from 1992 through 2007 were evaluated for inclusion in the study. Criteria for inclusion included age <15 days and a diagnosis of a cardiac defect based on necropsy or echocardiographic findings, or both. Signalment, presenting complaint, clinical and cardiac diagnoses, and outcome were recorded. In addition, physical examination findings including temperature, heart rate, respiratory rate, cardiac auscultation, mucous membrane color, and capillary refill time were recorded. Results of echocardiography, thoracic radiographs, blood gas analysis, and necropsy examinations were recorded. Recordings of echocardiographic examination were reviewed by 1 of the authors (M.D.K.) in 2 cases (foals 1 and 7) for which original echocardiographic reports could not be found. Foals were categorized based on primary cardiac defect into the following groups for the purpose of describing clinical and blood gas data: ventricular septal defect (VSD), tetralogy of Fallot (TOF), tricuspid valve atresia (TVA), combined TVA/VSD, TVA combined with pulmonary valve atresia (TVA/PVA), truncus arteriosus (TA), mitral valve dysplasia (MVD), total anomalous pulmonary venous return (TAPVR), and aortic valve stenosis.
A control population consisted of foals <15 days of age admitted with noncardiac problems during the same time period. This population was used to determine the breed and sex distribution of foals examined at the VMTH from January 1, 1992 through October 1, 2007.
Fischer's exact test was utilized to evaluate for breed and sex predilection. Odds ratios and 95% confidence intervals were calculated. Statistical significance was set at P < .05.
Twenty-two foals met the inclusion criteria. One foal was excluded as the echocardiogram examination was incomplete because of cardiac arrest and a necropsy was not performed. A 2nd foal was excluded because its sole cardiac necropsy findings included a patent foramen ovale (PFO) and patent ductus arteriosus (PDA), which were likely physiologic or secondary to sepsis. A 3rd foal was excluded as right atrial dilation was determined to be secondary to right ventricular myonecrosis rather than a congenital defect. A 4th foal with PDA and suspected tricuspid valve dysplasia on echocardiography was excluded because the dysplasia was not confirmed, it was considered clinically insignificant, and the foal was alive at the time of discharge from the hospital. This left 18 cases for inclusion in the study. Twelve additional horses with cardiac defects were found, but these were excluded because they were older than 14 days of age. The control group consisted of 1357 foals that were presented without a known cardiac defect.
Among the 18 foals that met the inclusion criteria, 15 had a necropsy performed, 16 had an echocardiogram exam, and 13 had both an echocardiogram and necropsy available for comparison. A complete list of cases is available in Table 1. Five foals had a single defect (VSD [n = 4] and aortic stenosis [n = 1]), while the other 13 foals had >1 cardiac defect.
|Foal||Breed||Echo||Necropsy||Outcome||Concurrent Diagnosis||Congenital Defects|
|1||Arab||2 VSD/OAo||VSD/TVA (PFO)||E|
patent urachus, bronchopneumonia, premature
|8||Arab||TOF||TOF||E||Dysmature, aspiration pneumonia|
|TA/VSD/ASD||E||Pulmonary edema/fibrosis, hepatic/splenic congestion||Ovarian cyst, flexural laxity|
|TVE||E||Dysmature, sepsis, colitis||Ovarian cyst|
|12||Mini||TOF||E||Pulmonary edema, atelectasis, hepatic vacuolation, premature||Flexural contracture|
|VSD/TVA (PFO)||E||Pulmonary edema, hepatic congestion, premature||Flexural contracture|
|E||Pulmonary fibrosis, hepatic congestion/fibrosis|
|E||Pneumonia, hepatitis, peritonitis||Meningocoele|
|17||Arab||VSD||TOF||D||Lung atelectasis, hepatic lipidosis, renal congestion, premature||Flexural laxity|
|E||Hepatic congestion, lung atelectasis/congestion|
Congenital cardiac defects associated with shunting (intra- or extracardiac) of blood were identified in 17/18 cases. Those identified on echocardiogram were confirmed with either color Doppler flow, bubblegram, or both. A VSD was diagnosed in 14 cases (5 of these were part of a TOF), including in 11 post-mortem studies and 11 echocardiograms (3 foals did not have necropsy examinations, 2 foals did not have echocardiograms, and 1 VSD was not observed on echocardiogram). Two of the VSDs were described as paramembranous and 2 as muscular. Five were described as “high” or “just distad to the aortic valve.” The size of the VSD was listed for only 4 cases (2 were 1 cm, 1 was 0.75 cm, and 1 was 2 cm in length). TOF was diagnosed in 5 of these foals; 2 of these were diagnosed on both echocardiographic and post-mortem examination, 1 each on echocardiography and postmortem alone, and in 1 the VSD component was identified on echocardiography while the entirety of the tetralogy was identified on necropsy (Fig 1).
Interatrial communications, including atrial septal defect (ASD) and PFO, were found in 8/18 cases, all in combination with other cardiac defects. Of these, 6/8 likely represented a physiologic PFO rather than an ASD. Seven were noted on postmortem; 5 had a shunt identified antemortem with echocardiography based on bubble contrast studies. PDA was found on post-mortem evaluation in 4 foals; the PDA was noted in only 1/4 echocardiogram studies. All foals with PDA had concurrent interatrial communication. It should be noted that all foals had complex cardiac anomalies in combination with the PDA.
A valvular defect was identified in 8/18 foals. Anomalies of the tricuspid valve, including atresia (Fig 2) and dysplasia, were diagnosed in 6 cases (2 with echocardiogram alone, 1 on postmortem alone, and 3 on both post-mortem exam and echocardiography) (Table 1). Other echocardiographic valvular defects included 1 case of MVD associated with dysplastic chordae tendinae and aortic insufficiency. A 2nd foal with TA and VSD had concurrent tricuspid valve regurgitation and aortic regurgitation. Valvular defects identified on pathology, but not observed on cardiac ultrasound, included 1 case of PVA and 1 case of aortic stenosis.
Congenital abnormalities involving the great vessels were identified in 5/18 cases. These included 1 foal each with TA and pulmonary artery hypoplasia on both echocardiography and pathology. Overriding aorta (OAo) was diagnosed in 2 foals, 1 with concurrent VSD and 1 with VSD and TVA. For purposes of grouping, the 1st foal was considered in the “VSD group” and the latter 1 in the “TVA/VSD group.” A 5th foal had TAPVR. In addition, 5 foals had OAo as part of the TOF complex.
Breed distribution of the study population included 7 Arabians (39%), 3 American miniature horses (17%), 2 Quarter horses (11%), 2 Morgans (11%), and 1 foal each of Thoroughbred, American Paint horse, Welsh pony, and Warmblood breed. Arabians were overrepresented compared with the general hospital neonatal population (OR 4.7, 95% CI = 1.8–12.4, P= .0035). Foals of Morgan breeding were also overrepresented (OR 11.2, 95% CI = 2.4–53, P= .02). Thoroughbreds were under-represented, although this was not statistically significant (P= .06). There were 8 males and 14 females. There was no significant sex (P= .2) predilection compared with the general hospital population. Age ranged from <24 hours to 14 days of age, with an average age of 3.9 ± 4.1 days (median = 2).
Reason for Presentation or Referral to the VMTH
Foals were presented to the VMTH for a variety of reasons, which are listed below by group.
Of 5 foals with VSD (including 1 foal with concurrent OAo), 1 foal each had sepsis, prematurity, omphalophlebitis, and neurologic signs (the foal with concurrent meningocoele) on initial examination. One foal was dead on arrival to the VMTH. Of 5 foals with TOF, 2 foals were examined for prematurity and weakness, and 2 for cardiac defects/heart failure based on the presence of cyanosis, tachycardia, and respiratory distress noted by the referring veterinarian. One was examined because of congenital anomalies in other body systems. Of 3 foals with both TVA and VSD, 1 foal each was examined for pneumonia, nonspecific illness, and mare rejection.
The single foal with TVA and PVA was presented with suspected neonatal maladjustment syndrome. The foals with TA and TAPVR were referred for weakness. The foal with aortic stenosis was presented with nonspecific signs of illness, and the foal with MVD presented with weakness following induced parturition.
The findings on initial examination are reported by primary defect in Table 2. Physical examination findings were available in 17/18 cases (1 foal with a VSD was dead on arrival). On initial examination 82% of the foals (14/17) were tachycardic (>100 beats/min). One foal with a VSD was bradycardic (60 beats/min). Two foals had normal heart rates (80–100 beats/min), including 1 with a VSD and 1 with TAPVR.
|VSD (n = 5)|
(includes 1 foal with concurrent OAo)
Arabian (1) Warmblood (1) Welsh (1) Morgan(1)
Not performed (3)
|TOF (n = 5)||Arabian (2)|
American miniature (2)
Not performed (2)
|TVA/VSD (n = 3)||Arabian (1)|
|Cyanotic (3)||Systolic (2)|
|Not performed (3)|
|TVA/PVA (n = 1)||Arabian||120||56||Cyanotic||Continuous/LHB||4/6||Interstitial|
|TA (n = 1)||Arabian||158||72||Muddy||Continuous/LHB||4/6||No significant findings|
|TAPVR (n = 1)||Morgan||90||60||Cyanotic||Holosystolic/RHB||5/6||Cardiomegaly|
|AS (n = 1)||Paint||120||54||Pink||Continuous/PMI not recorded||n/r||Interstitial|
|MVD (n = 1)||Arabian||180||24||Pink||Continuous/PMI not recorded||5/6||Alveolar|
All foals were tachypneic (>24 breaths/min) on admission. Seven foals had cyanotic mucous membranes and 7 had pink mucous membranes. Those with cyanotic mucous membranes had intracardiac shunts, including 6 with right to left shunting (4 TVA, 2 TOF) and 1 with left to right shunting (TAPVR). However, with TAPVR there is always right-to-left shunting through interatrial communications as well, so in fact the cyanosis is caused by right-to-left shunting in this case also. Five of these foals with cyanotic mucous membranes had results of arterial blood gas analysis available, and in all 5 severe hypoxemia (PaO2 < 50 mmHg) was confirmed. One foal (foal 6) with a VSD was also found to be hypoxemic; however, this foal had concurrent pneumonia and pulmonary hypertension with a left to right shunt through the VSD. The hypoxemia in this foal was attributed to the lung lesions.
All foals had a heart murmur graded ≥3/6. Murmurs were described as holosystolic with point of maximum intensity (PMI) at the right heart base in foals with VSD and systolic with PMI at the left heart base (LHB) in foals with TOF or TVA. Thoracic radiographs were abnormal in 9/10 foals, with marked variability in findings among foals. Results of cardiac auscultation and thoracic radiography are presented in Table 2.
Blood Gas Analysis
Results of arterial blood gas analysis were available for 9/18 foals. The results are summarized by primary cardiac defect in Table 3. Eight of the 9 foals had PaO2≤ 45 mmHg indicating severe hypoxemia. Only 2/9 (1 TVA and 1 VSD) had arterial sample results available shortly after intranasal oxygen therapy was initiated; the foal with TVA had confirmed right to left shunting of blood.
|VSD (n = 1)||7.412||7.407||40.2||43.4||47.7||42.7|
|TOF (n = 3)||7.448||21||24.2|
|TAPVR (n = 1)||7.25||46||29.5|
|MVD (n = 1)||7.458||59.8||40.2|
Nine of the 18 foals had concurrent congenital anomalies of other systems (Table 1). Six foals were also classified as premature or dysmature based on gestational age, clinical examination, or both. Other concurrent diagnoses were predominantly based on necropsy findings, except for 3 foals that were discharged.
One foal was dead on arrival, 1 died during hospitalization, and 13 foals were euthanized based upon poor prognosis and/or poor response to medical therapy. The overall mortality was 83% (15/18). Of those euthanized, 7/13 were euthanized due to the presence of the cardiac abnormality. The remaining 6 foals were euthanized for a variety of reasons, including poor prognosis or nonresponse to therapy for other concurrent diseases.
Three foals survived to discharge and of these, 2 had died at the time of follow-up with the owners. Foal 5, with TOF, died suddenly at 2.5 months of age. Foal 3 was diagnosed with a VSD and was euthanized at 8 years of age after becoming recumbent secondary to suspected osteoarthritis. Foal 10 was originally diagnosed with a VSD and aortic insufficiency. He was reevaluated as a yearling and the VSD had apparently closed at that time but the aortic insufficiency had progressed on echocardiographic examination. This horse is still alive at 10 years of age and is performing as a level 3 dressage horse.
An association between the diagnosis of a congenital cardiac defect and Arabian breeding was confirmed in this study population. Pedigrees of these animals were not reviewed; however, the overrepresentation of Arabians with heart defects as compared with the general hospital neonatal foal population is suggestive of a genetic susceptibility within this breed. The small number of foals in the present study warrants caution in extrapolating these results to Arabian horses in other regions. However, in addition to the 7 Arabian foals from this study, 18 of 47 foals (38%) from previous case reports of cardiac defects were also Arabians.1,22–33 Other previously reported breeds included 9 Thoroughbreds, 2 Welsh ponies, 1 American miniature horse, and 1 Morgan foal. The association between heart abnormalities and Morgan horses identified in our report must be interpreted with caution due to the low number of cases, but warrants further investigation.
Based on the results of this study, congenital cardiac defects of equine neonates are often complicated by multiple cardiac abnormalities. Of the defects identified, VSD was most common, being present in 14/18 (78%) of the foals. VSD independent of TOF was present in 50% of foals and TOF was found in 28%. The other common defect was TVA (22%). Interatrial communications were identified in 44% of cases (8/18) and these were confirmed with shunting of blood identified on echocardiogram in 5/8 of these cases. PDA was identified in 4 foals but only 1 foal was greater than 4 days of age at the time of diagnosis on post-mortem examination, which makes it difficult to know whether these were physiologic or pathologic PDAs. It should be noted that there were some discrepancies between echocardiographic and post-mortem diagnoses (Table 1). These discrepancies might reflect experience level of echocardiographer or pathologist.
Limiting the age range to the neonatal period (< 15 days) likely influenced the high case fatality rate in this study by selecting for the most severe defects. An uncomplicated VSD is reported to be the most commonly encountered congenital cardiac defect in horses,2,3,28,35 however, many of these remain undiagnosed until the animal reaches training age and demonstrates exercise intolerance or a heart murmur. In the present study, a VSD was the sole defect found in 4 foals. In the other 10 foals, a VSD was associated with TOF (5), TVA (3), TA (1), and OAo (1), and was likely essential for the foals' survival beyond birth in many cases. Also, 17/18 foals (94%) had defects that resulted in intra- or extracardiac shunting of blood (9/17 were right-to-left), which likely contributed to the case fatality rate of 83%. The severe hypoxemia in the 9 foals in which arterial blood gas tensions were recorded and the poor response to oxygen insufflation in 2 of the foals with follow-up blood gas analysis is consistent with severe right-to-left shunting of blood.
The high prevalence of intracardiac shunts in the present study has not been reported previously in horses. The incidence of TOF in the present study (28%) was much higher than that reported in a necropsy study of 32 equine neonatal hearts in which only 1 foal had TOF.2 The reason for this difference is unknown, but might be due to the population of horses or nature of referral practice. Interatrial communication was present in 44% of foals in the current study, which is consistent with previous reports regarding ASD and PFO.2
The presence of PFO or PDA in this study must be interpreted with caution. Physiologic or functional closure of the ductus arteriosus occurs in the equine neonate between 48 and 72 hours of age,34 whereas complete anatomic closure could be delayed until 4 days of age.3,15 Physiologic or functional closure of the foramen ovale occurs immediately postpartum; however, anatomic closure does not occur until 15 days to 9 weeks of age.3 Therefore, a diagnosis of either a PFO or PDA on postmortem not confirmed with identification of shunting on echocardiogram might not be clinically significant. Alternatively, echocardiographic identification of shunting through a PFO or PDA can be difficult and it is possible that it might not have been identified in some cases. These defects could have closed had the foals survived and therefore might have been clinically insignificant. However, the PFO and PDAs present with complex defects were likely essential to the foals' survival beyond birth.
All of the foals in this study had a murmur graded as ≥3/6. This underscores the importance of pursuing the origin of moderately loud-to-loud murmurs in neonatal foals. Auscultable murmurs were ≥grade 3/6 (100%), systolic (65%) and had a point of maximum intensity at the LHB most commonly (50%). Tachycardia was common; however, heart rate was variable among foals with VSD, including brady-, normo-, and tachycardia. Tachypnea (>24 bpm) was present in all foals, while 9/17 (53%) were presented with severe tachypnea (≥60 bpm). There was an association between mucous membrane color and the broad category of shunting defects, as all of the foals presented with cyanotic membranes had a right-to-left shunting defect. In addition, all of these foals were markedly hypoxemic and 2 foals with follow-up analysis of blood gas tension showed poor response to oxygen insufflation. Of the available results for arterial blood gas analysis, 89% of foals were markedly hypoxemic. Thoracic radiographic findings were variable, but abnormal in most foals.
Interestingly, 6/18 foals were determined to be premature or dysmature based on gestational age or physical examination. Premature or dysmature foals should be examined closely for cardiac defects. One third of the foals also had multiple concurrent congenital abnormalities. These noncardiac anomalies were heterogeneous, which highlights the importance of closely evaluating the heart for defects in foals presenting with other congenital abnormalities. Cardiac defects represented 3.5% of all congenital defects in equine neonates in a study of post-mortem results, which is similar to the findings of this retrospective study where congenital cardiac defects were identified in 1.6% of the general hospital population of equine neonates.2
There appears to be an association between Arabian breed and the occurrence of congenital cardiac defects. The presence of a loud murmur, cyanotic membranes, and tachycardia or tachypnea in a neonatal foal should warrant close attention to the heart structure for the possibility of a congenital defect. In addition, the identification of a cardiac defect in the equine neonate carries a poor prognosis when associated with a right to left shunt. Horses with VSD can survive long-term and at least 1 horse was successfully performing as an athlete.
Dr Bill Thomas for providing images of congenital cardiac defects.