Introduction
As veterinary care and welfare for nondomestic hoofstock housed in modern zoological institutions continues to improve, there are greater numbers of geriatric populations under human care. It is now common to have captive animals that surpass the life expectancy for the species in the wild, although it is also important to consider that many of these life expectancies continue to change as geriatric animals age at zoological institutions.1,2 With this evolution of captive species, improved veterinary care is paramount to increased longevity with higher standards for diagnostic and therapeutic options.
Electrocardiography and thoracic radiography are 2 noninvasive diagnostic tools for evaluation of cardiothoracic morphology and cardiac electrical activity. These tools are also useful to identify specific degenerative changes related to the cardiac function. There is a growing body of literature describing measurements of ECG and thoracic imaging parameters in zoo species.3–7 Ruminants, especially cervids, are considered high risk species for chemical immobilization. This is due to their susceptibility to capture myopathy during stressful procedures. As such, there is little information related to proper cardiac assessment in cervids.7
Because cervids in captive settings have longer life expectancies than wild cervids, geriatric cervid populations in zoo settings are increasingly common. Some of the most common degenerative changes related to aging in domestic ruminants are related to alterations in the contractility and cardiac electrical activity of the heart.8 To the authors ’ knowledge, there are no published studies evaluating these changes in nondomestic ruminants. Furthermore, ruminant anesthesia is often performed in a free-range setting, so acquiring a conventional ECG tracing is not always feasible. A suitable alternative for this task, a smartphone-based ECG (aECG) is an inexpensive device that has been validated for the routine, portable monitoring of the cardiac rate and rhythm in other species.5–7
The purpose of this study was to describe cardiothoracic qualitative and quantitative values from radiographic and electrocardiographic evaluations performed in a population of geriatric Sika deer (Cervus nippon). Additionally, a comparison of 2 ECG devices—6-lead ECG (sECG) and aECG—was performed. We hypothesized that there would be no significant difference in median heart rate, rhythm, amplitude, and duration of the measured ECGs between the sECG and aECG in this population.
Materials and Methods
Animals
This study was conducted using 10 geriatric Sika deer (9 females and 1 male), housed together in a multispecies exhibit in a private safari operation located in Winston, Oregon with an elevation of 545 feet. Median age was 17.5 years (range, 15 to 29 years), and median body weight was 55.25 kg (range, 41.8 to 65.5 kg). The study was conducted during routinary annual physical examinations under chemical immobilization. No respiratory or cardiac abnormalities were auscultated during physical examination. Food was not withheld from deer prior to study procedures, as animals were not separated from the herd prior to anesthetic induction.
Anesthesia
For chemical immobilization, Sika deer were administered a combination of tiletamine-zolazepam (1.3 mg/kg, IM) and xylazine (1 mg/kg, IM) by means of a remote drug delivery system (X-Caliber; Pneu-Dart), and a 2-mL disposable dart (1-inch gel collar P type; Pneu-Dart). Immediately after immobilization, all deer were placed in right lateral recumbency, and a 2.9 cm, 18-gauge IV catheter was percutaneously placed in the left lateral saphenous vein. The deer were then transported to the safari hospital and weighed. A blood sample was collected from the left jugular vein for hematologic and biochemical analysis. Heart rate, respiratory rate, esophageal temperature, indirect blood pressure, capnography results, and relative blood oxygen saturation were recorded using a veterinary multiparameter device (Datascope Spectrum) every 5 to 10 minutes throughout the anesthetic event. A 7.5-mm cuffed endotracheal tube was placed to avoid atelectasis due to the prolonged lateral recumbency. Oxygen flow was initiated at 2 L/min with a circular rebreathing circuit.
Radiography
All Sika deer were placed in right lateral recumbency with hyperextension of the thoracic limbs to allow proper radiographic positioning of the thorax. Lateral projections of the thorax were taken with the dorsal rib heads superimposed during full inspiration. A focal distance of 100 cm was utilized, with exposure settings of 90 kilovolts (peak) and 4.0 mA.
The maximum thoracic thickness of each deer was determined with calipers. The radiograph cassette (35.6 X 43.2 cm) was horizontally positioned, allowing the complete visualization of the vertebral column and sternum. A complete radiographic examination of the thorax was denoted by inclusion of the scapula and the first ribs cranially, the thoracic vertebral column from C7 to T12, the sternum, the humeral-radioulnar joint (ventrally), and the diaphragm caudally. An orthogonal view was obtained with the deer in ventrodorsal recumbency with hyperextension of the thoracic and pelvic limbs, utilizing the same anatomic structures to demark the exposed regions. The orthogonal radiographs had identical exposure settings to those of the right lateral views.3,9,10
Direct measurements were made from the radiographs utilizing a DICOM software viewer (Osirix Lite; Pixmeo SARL) and measurement tool. Absolute measurements of structures of interest were compared with measurements of other consistent anatomic structures, allowing standardization within deer, independent of body size. Measurements were obtained by described techniques in other ruminant and nonruminant mammalian species. In the right lateral projection, the cardiac height (base to apex) and cardiac width (cranial to caudal) were measured (Figure 1).3,9,10 The height of the heart was determined from the ventral portion of the mainstem bronchi to the apex of the heart. The width was determined from the junction of the ventral margin of caudal vena cava with the right atrium to the cranial cardiac margin, perpendicular to the height measurement.3,9,10
Representative right lateral radiographic image showing thoracic dimensions measured in anesthetized Sika deer (Cervus nippon). 1 = Cardiac height. 2 = Cardiac width. 3 = Cardiosternal contact. 4 = Thoracic height. 5 = Cardiophrenic contact. 6 = Tracheal divergence angle. 7 = Distance of T3 to T5.
Citation: American Journal of Veterinary Research 83, 2; 10.2460/ajvr.21.08.0128
Representative right lateral radiographic image showing thoracic dimensions measured in anesthetized Sika deer (Cervus nippon). 1 = Cardiac height. 2 = Cardiac width. 3 = Cardiosternal contact. 4 = Thoracic height. 5 = Cardiophrenic contact. 6 = Tracheal divergence angle. 7 = Distance of T3 to T5.
Citation: American Journal of Veterinary Research 83, 2; 10.2460/ajvr.21.08.0128
Representative right lateral radiographic image showing thoracic dimensions measured in anesthetized Sika deer (Cervus nippon). 1 = Cardiac height. 2 = Cardiac width. 3 = Cardiosternal contact. 4 = Thoracic height. 5 = Cardiophrenic contact. 6 = Tracheal divergence angle. 7 = Distance of T3 to T5.
Citation: American Journal of Veterinary Research 83, 2; 10.2460/ajvr.21.08.0128
These values were compared to the total height of the thorax, which was determined from the ventral apex of the heart to the ventral margin of the mid-caudal endplate of T5, and the distance between T3 and T5, which was determined by the mid-cranial endplate of T3 to the mid-caudal endplate of T5.
The additional criteria measured to assess the cardiac size included the intercostal spaces occupied by the width of the heart, tracheal angle of divergence from the thoracic vertebral column, and degree of cardiophrenic and cardiosternal contact.3,9,10 The tracheal angle was determined by drawing straight lines along the dorsal border of the trachea and the ventral margin of the cranial thoracic vertebrae and measuring the resultant angle of intersection cranially. Cardiophrenic contact was determined by measuring from the apex of the heart to the most dorsal point of interface of the cardiac silhouette with the diaphragm. Cardiosternal contact was determined by measuring how many sternebrae were in contact with the heart. Qualitative assessment of cardiac shape and orientation was recorded.3
A modified vertebral heart score (VHS) previously used in other ruminant species was proposed to attempt to evaluate the heart size and to describe these values in Sika deer. Since the carina is not readily apparent on normal thoracic radiographs in this species, the ventral margin of the left mainstem bronchus was used for the longitudinal measurement of the heart.3,11
After completion of radiography, anesthesia with isoflurane was initiated. Anesthesia was maintained with isoflurane at 1 to 3%. Supportive fluid therapy using lactated Ringer solution (10 mL/kg/h, IV) was initiated at this point and for the remainder of the procedure.
Electrocardiography
Six-lead ECG (Model 1700A; Hewlett Packard) recordings were obtained with anesthetized deer in right lateral recumbency. Thoracic and pelvic limbs were positioned perpendicular to the long axis of the body, and electrodes were placed proximal to the elbow joint in the forelimbs and proximal to the stifle joint in the pelvic limbs.5,7 Electrodes were attached to the skin using isopropyl alcohol to enhance signal transmission. Recordings were acquired using a paper speed of 50 mm/s, sensitivity of 20 mm/mV, and a duration of 30 seconds.
Simultaneously a bipolar aECG (AliveCor) recording with the same speed, amplitude and duration, was recorded with the smartphone placed over the left lateroventral thorax, centered over the heart. An application downloaded onto the smartphone transformed the electrical signal into an ECG. The sECG recordings were automatically digitalized by the device and stored as a PDF file on the smartphone.5,6
ECG analysis
The aECG recordings were digitally stored and later printed with a paper speed of 50 mm/s and amplitude of 20 mm/mV; these recordings were analyzed manually in conjunction with the printed sECG recordings. The RR intervals were measured using calipers.
Both the aECG and sECG recordings were examined to identify identical complexes and RR intervals. Once matched, 20 consecutive cardiac cycles were measured and evaluated. The instantaneous heart rate was obtained by counting the number of small boxes (1 mm = 0.02 seconds) in 1 R-R interval, and then dividing the constant of 3,000 by the total number of small boxes. A minimum and maximum instantaneous heart rate was also calculated. The underlying cardiac rhythm was determined by careful inspection of all complexes in lead II. Standard ECG measurements were performed, including P wave duration and amplitude (s and mV); QRS duration and amplitude (s and mV); PR and QT intervals (s); and T wave direction and amplitude (mV). All measurements of complexes and intervals were performed using lead II.5–8 The mean electrical axis was calculated using the QRS complexes of the lead perpendicular to the most isoelectric lead for the sECG; however, this parameter was not calculated for the aECG due to a single lead being present (Supplementary Figure S1).
Statistical analysis
Statistical analysis was performed using 2 commercially available software programs (ToolPak®; Excel 2019® Microsoft-Office 2019®; X real stats ®; Real statistics®; Charles Zaionts). Descriptive statistics were performed for the radiographic measurements. Significant differences between the sECG and the aECG were assessed using the Wilcoxon signed rank test for nonparametric matched data with a significance set at P < 0.05.
Results
Animals
All Sika deer were considered in good body condition based on the physical examination findings, as well as with normal hematological and biochemical results. Chemical immobilization using the protocol described above resulted in rapid anesthetic induction (< 10 minutes) with adequate muscle relaxation, and smooth recovery from anesthesia.
Radiology
Right lateral and ventrodorsal views were obtained for all deer. All deer had mild to moderate spondylosis deformans noted in the thoracic vertebral column. Occasional osteophytes were seen in the ventral portion of the thoracic vertebral bodies, with no severe bone remodeling or ankylosing spondylitis that could potentially interfere with the accurate identification of the anatomic landmarks needed for this study. No pathologic abnormalities associated with cardiac morphology were noted involving the thoracic cavity. Radiographs showed normal anatomic landmarks as previously described, and no evidence of cardiac or pulmonary disease was noted.
Values for cardiac and thoracic parameters as radiographically measured were summarized (Table 1). Data dispersion was low, as inferred by comparing mean values with their corresponding SDs.
Descriptive statistics for cardiac and thoracic parameters as radiographically measured in 10 chemically immobilized, geriatric Sika deer (Cervus nippon).
| Parameter | Mean ± SD | 95% CI* | Median | Range |
|---|---|---|---|---|
| Cardiac height (mm) | 170 ± 12 | 163–178 | 170 | 153–193 |
| Cardiac width (mm) | 134 ± 6 | 130–138 | 135 | 122–146 |
| Thoracic height (mm) | 235 ± 10 | 228–241 | 238 | 217–245 |
| Distance between T3 and T5 (mm) | 106 ± 7 | 101–110 | 104 | 96–123 |
| Tracheal angle (°) | 18 ± 4 | 16–21 | 19 | 13–25 |
| Cardiophrenic contact (mm) | 78 ± 11 | 71–85 | 75 | 60–98 |
| Cardiosternal contact (mm) | 98 ± 19 | 86–110 | 99 | 69–124 |
| No. of sternebrae in contact with heart (mm) | 3 ± 0 | 2–3 | 3 | 2–3 |
| Size of 1 vertebra (mm)† | 35 ± 2 | 34–37 | 35 | 32–41 |
| Vertebral heart score | 8.6 ± 0.5 | 8.4–8.9 | 8.7 | 7.6–9.2 |
Values represent the 95% CI of the mean value. †Values represent the distance between T3 and T5, divided by 3.
Electrocardiography
All Sika deer had a normal sinus rhythm, with no pathologic arrhythmias noted. Results of ECG were summarized for the sECG and aECG (Table 2). The Wilcoxon signed rank test confirmed a lack of significant differences between the 2 devices in maximum HR, QRS duration and amplitude, and T wave amplitude. However, there was a significant difference between devices in minimum heart rate (P = 0.001), P wave duration (P = 0.007) and amplitude (P = 0.004), PR interval (P = 0.02), and QT interval (P = 0.04).
Descriptive statistics for values measured by 6-lead ECG (sECG) smartphone ECG app (aECG) in the deer of Table 1.
| Parameter | Device | Mean | Median | Range |
|---|---|---|---|---|
| Maximum HR (bpm) | sECG | 55 ± 8 | 53 | 44 to 66 |
| aECG | 56 ± 8 | 53 | 48 to 67 | |
| Minimum HR (bpm) | sECG | 49 ± 10 | 46a | 36 to 65 |
| aECG | 51 ± 10 | 48b | 37 to 66 | |
| P wave duration (s) | sECG | 0.05 ± 0.01 | 0.05a | 0.04 to 0.06 |
| aECG | 0.03 ± 0.01 | 0.03b | 0.02 to 0.04 | |
| P wave amplitude (mV) | sECG | 0.28 ± 0.04 | 0.30a | 0.23 to 0.34 |
| aECG | 0.10 ± 0.03 | 0.09b | 0.07 to 0.15 | |
| QRS duration (s) | sECG | 0.07 ± 0.01 | 0.07 | 0.06 to 0.08 |
| aECG | 0.07 ± 0.01 | 0.07 | 0.05 to 0.08 | |
| QRS amplitude (mV) | sECG | 0.88 ± 0.48 | 0.77 | 0.24 to 1.66 |
| aECG | 0.58 ± 0.28 | 0.58 | 0.23 to 1.07 | |
| PR interval (s) | sECG | 0.15 ± 0.03 | 0.15a | 0.08 to 0.18 |
| aECG | 0.12 ± 0.01 | 0.12b | 0.10 to 0.14 | |
| QT interval (s) | sECG | 0.39 ± 0.05 | 0.38a | 0.33 to 0.49 |
| aECG | 0.30 ± 0.14 | 0.37b | 0.04 to 0.43 | |
| T wave amplitude (mV) | sECG | 0.58 ± 0.40 | 0.45 | 0.12 to 1.30 |
| aECG | 0.44 ± 0.27 | 0.35 | 0.08 to 0.84 | |
| Mean electrical axis (°) | sECG | –66.0 ± 85.8 | –105 | –120 to 120 |
a,bValues with different superscript letters differ significantly (P < 0.035).
bpm = Beats per minute. HR = Heart rate.
Discussion
Our study showed that the aECG, an affordable option for performance of ECG, was a feasible method for evaluating heart rate and rhythm in Sika deer during field immobilization protocols. However, as discussed in previous publications5,6,12 concerning other species, the aECG was not a direct substitute for assessment of ECG waveforms, amplitudes, durations, or timing intervals, although heart rate and rhythm could be routinely identified using the aECG in the animals of this study.
Furthermore, this study also provided foundational data for thoracic radiography in geriatric captive deer, suggesting that this noninvasive diagnostic technique may be used to assess cardiothoracic morphology during annual health examinations of captive deer. Although the gold standard for assessing cardiac function is echocardiography, thoracic radiography may be a good initial screening tool in deer because radiography does not require highly specialized personnel or ultrasound equipment.
Radiographic evaluation of the thorax generally requires 2 orthogonal views to properly assess relevant cardiothoracic structures. In the present study, right lateral and ventrodorsal views were acquired, although only the right lateral view was used for measurement purposes. Normally in ruminant species, obtaining a right lateral view of the thorax is less complicated than a ventrodorsal or dorsoventral view. However, the limitations of a single lateral radiographic view include incomplete pulmonary assessment and difficulty in differentiating pulmonary from mediastinal abnormalities due to summation in the lateral view.3,9,10
For the deer of the present study, the mean ± SD VHS of 8.6 ± 0.5 was similar to that of other ruminants, such as female domestic Bergamasca sheep (8.99 ± 0.27).9 Our values were slightly less than those reported in alpaca crias10 (9.36 ± 0.65) and other sheep breeds13 (10.23 ± 0.43). The mean tracheal angulation of 18 ± 4° in Sika deer was slightly greater than the mean angle reported for domestic llamas3 (14.4 ± 2.0°) and alpaca crias10 (14.24 ± 3.57°). The maximum measurement obtained in our deer was similar to the maximum reported with other breeds of sheep13 (22.50°). For the degree of cardiophrenic contact observed, 6 of the 10 deer had a value of 70 to 80 mm, which is similar to other breeds of sheep13 (mean ± SD overlap, 66.50 ± 1.67 mm). Lastly the cardiosternal contact obtained was of 3 sternebrae, which is similar to the reported contact in sheep and llamas.3,13
With respect to ECG measurements, our study showed that the aECG was able to identify and record normal electrocardiographic waveforms in all Sika deer. Although there were significant differences between the aECG and sECG in ECG waveform components, there was no difference in rhythm diagnosis between methods. Thus, the aECG was useful to calculate overall heart rate and rhythm but may not been an equivalent substitute for evaluating duration and amplitude of ECG waveforms, compared with sECG.5,6,12
The significant ECG differences that we found between ECG devices were relatively minor and unlikely to be clinically important. For example, the P wave duration was slightly longer, and amplitude slightly higher, on the sECG than on the aECG. The duration and amplitude of the P wave and the PR interval generally reflects atrial mass and atrioventricular conductivity,8 so the aECG may not be a direct substitute for changes in P wave morphology that may reflect atrial pathology. The PR interval and the QT interval on the sECG were statistically slightly longer than on the aECG, which likely reflected inaccuracy of the aECG, as the QT interval is also dependent on heart rate.
A previous study7 described various morphological ECG findings in Roe deer (Capreolus capreolus) comprised of negative P waves, biphasic or negative T waves, and biphasic QRS complexes. These findings are partially explained by the Purkinje fiber orientation and distribution in ruminant species. The distribution of Purkinje cells in the ruminant heart is diffuse compared with non-ruminant species, and it is composed by Purkinje cells in the forms of clusters and single cells in the lateral and mural myocardium. Therefore, the QRS morphology can be quite variable, and its duration will normally be shorter than in other species.14,15 Other species with more predictable QRS complexes have different Purkinje fiber distributions, creating a pattern of depolarization with consistent QRS morphology. For instance, the mean electrical axis range for the Sika deer of the present study was quite variable at –120° to 120°. Findings of another study5 in water buffalo (Bubalus bubalis) calves suggest a similar mean electrical axis. The mean electrical axis is net vector determined by activation of the interventricular septum and the ventricular free walls. It has been stated that thoracic limb leads, and heart ’s orientation and location could also account for variations in directions of the mean QRS complex. So, species-specific cardiac morphology can alter this complex morphology as well.5,7,12,14–16
Some of the other values obtained with the sECG such as P wave duration, QRS duration, PR interval, and QT interval were either very similar to or slightly different from those reported in Roe deer and domestic ruminant species.7,8,17 The mean ± SD T wave amplitude recorded in our Sika deer was 0.58 ± 0.40 mV, which was nearly identical to reported values for Roe deer (0.59 ± 0.33 mV) in the base-apex measurement performed in that study,7 but slightly higher than the amplitude obtained with the lead II standard bipolar and augmented unipolar limb leads in the same Roe deer (0.20 ± 0.79 mV) and in geriatric cattle8 (0.34 ± 0.01 mV) and Chios sheep17 (0.12 ± 0.050 mV). Such differences when comparing base-apex ECG to the standard bipolar limb leads in both deer and Chios sheep should be taken into consideration when measuring and comparing cardiac electrical activity in ruminants.7,17 Traditionally, the base-apex lead configuration is the most useful in measuring conduction times in ruminant species because the origins and terminations of deflections could be identified easily. We did not have any issues with the lead configuration used in our study, and deflections were easily identified.
It is important to mention that the literature regarding ECG in aging ruminants is scarce, and typically the adult or geriatric populations described in these studies8,17 do not exceed 8 years of age. The Roe deer in the previous study7 were described as > 2 or < 2 years of age. It can be suggested, as with domestic geriatric cattle8 and Chios sheep,17 that with the aging process, the heart becomes larger and the surface of cardiac electrical activity increases. But further research is needed in several cervid species of different ages to assess the validity of this suggestion.
The present study had several limitations. First, most Sika deer were 15 to 19 years old, with just 3 animals > 25 years old. Thus, the generalizability of our findings to the geriatric Sika deer population in general is unclear. Also, our deer had no echocardiograms to confirm they did not have clinically relevant cardiovascular disease; however, no audible murmurs were identified, which generally are associated with significant valvular or myocardial dysfunction. We also did not compare the cardiac measurements on radiographs or ECG to a younger population, so we could not determine what changes are related to the normal aging process.
In conclusion, although very limited information is available to compare radiographic and ECG measurements in domestic and nondomestic ruminant species, the right lateral thoracic radiographic view was found to be suitable for evaluating cardiothoracic morphology in Sika deer. These geriatric deer had no morphologic abnormalities related to cardiac structure or electrical activity. We found the evaluated aECG to be a feasible, cost-effective option for monitoring heart rate and rhythm. Nevertheless, specific waveforms on the aECG should be interpreted with caution and confirmed or refuted by sECG.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
The authors declare there were no conflicts of interest.
The authors thank the Carlson Veterinary College at Oregon State University for donating the 6-lead ECG used in this research and Alejandra Jassi and Marina Garcia for performing the statistical analysis.
References
- 1. ↑
Krebs BL, Marrin D, Phelps M, Krol L, Watters JV. Managing aged animals in zoos to promote positive welfare: a review and future directions. Animals (Basel). 2018;8(7):2–22. .
- 2. ↑
Kohler IV, Preston SH, Bingaman L. Comparative mortality levels among selected species of captive animals. Demogr Res. 2006;15:413–434. .
- 3. ↑
Mattoon JS, Gerros TC, Brimacombe M. Thoracic radiographic appearance in the normal llama. Vet Radiol Ultrasound. 2001;42(1):28–37. .
- 4.
Schumacher J, Snyder P, Citino SB, Bennett RA, Dvorak LA. Radiographic and electrocardiographic evaluation of cardiac morphology and function in captive cheetahs (Acinonyx jubatus). J Zoo Wildl Med. 2003;34(4):357–363. .
- 5. ↑
Smith J, Ward J, Urbano T, Mueller M. Use of AliveCor Heart Monitor for heart rate and rhythm evaluation in dairy water buffalo calves (Bubalis bubalis). J Dairy Vet Anim. 2016;4(2):1–4. .
- 6. ↑
Yaw TJ, Kraus MS, Ginsburg A, Clayton LA, Hadfield CA, Gelzer AR. Comparison of a smartphone-based electrocardiogram device with a standard six-lead electrocardiogram in the Atlantic bottlenose dolphin (Tursiops truncatus). J Zoo Wildl Med. 2018;49(3):689–695. .
- 7. ↑
Santamarina G, Espino L, Suarez ML. Electrocardiographic parameters of free-ranging roe deer (Capreolus capreolus). J Zoo Wildl Med. 2001;32(4):441–446. .
- 8. ↑
Sudhakara Reddy B, Sivajothi S. Electrocardiographic parameters of normal dairy cows during different ages. J Veter Sci Med. 2016;4(1):1–4. https://avensonline.org/wp-content/uploads/JVSM-2325–4645–04–0023.pdf.
- 9. ↑
Babicsak VR, Alves LS, Tsunemi MH, Vulcano LC. Radiographic measurements related with the cardiac size in young female Bergamasca sheep. Pesqui Vet Bras. 2017;37(12):1526–1530. .
- 10. ↑
Nelson NC, Mattoon JS, Anderson DE. Radiographic appearance of the thorax of clinically healthy normal alpaca crias. Am J Vet Res. 2011;72(11):1439–1448. .
- 11. ↑
Buchanan JW, Bücheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc. 1995;206(2):194–199. https://pubmed.ncbi.nlm.nih.gov/7751220/.
- 12. ↑
Smith J, Heller M, Smith F, et al. Use of an AliveCor Heart Monitor for heart rate and rhythm evaluation in domestic goats. In: Proceedings of the American College of Veterinary Internal Medicine Research Forum. American College of Veterinary Internal Medicine; 2016;30:1516.
- 13. ↑
Makungu M. Thoracic radiographic anatomy in sheep. In: Proceedings of the Tanzania Veterinary Association. Tanzania Veterinary Association; 2017;32:129–135.
- 14. ↑
Ryu S, Yamamoto S, Andersen CR, Nakazawa K, Miyake F, James TN. Intramural Purkinje cell network of sheep ventricles as the terminal pathway of conduction system. Anat Rec (Hoboken). 2009;292(1):12–22. .
- 15. ↑
Ansari A, Yen Ho S, Anderson RH. Distribution of the Purkinje fibers in the sheep heart. Anat Rec. 1999;254(1):92–97.
- 16. ↑
Smith CR, Hamlin RL, Crocker HD. Comparative electrocardiography. Ann N Y Acad Sci. 1965;127(1):155–169.
- 17. ↑
Chalmeh A, Aktar IS, Zarei MH, Badkoubeh M. Electrocardiographic indices of clinically healthy Chios sheep. Vet Sci Dev. 2015;5(2):99–102. .
