Elasmobranchs are cartilaginous, aquatic vertebrates belonging to the Chondrichthyes class of fish.1 Within the subclass Elasmobranchii, there are approximately 400 species of sharks and 500 species of rays, skates, and sawfish.1 Given the size of the taxonomic group and large variation in the ecological niche of each species, dramatic differences in anatomy and physiology exist between species. There is a dearth of literature describing normal cardiac anatomy and physiology in stingrays. Instead, most of the research in elasmobranchs has focused on a select few species of shark.2–5 Briefly, elasmobranchs have a 2-chambered heart (atrium and ventricle). The atrium receives blood from an accessory chamber, called the sinus venosus, which collects blood from systemic veins. Blood moves from the atrium, through the atrioventricular valve into the ventricle, which pumps blood into the muscular conus arteriosus (conus) for distribution to the gills.6 There is also a lack of published information on cardiac pathology in elasmobranchs. Case reports describing dilated cardiomyopathy,6,7 parasitic,8,9 nutritional,10,11 capture myopathy with cardiac involvement,12 and iatrogenic disease processes affecting the cardiovascular system (gas bubble disease,13 tattoo ink emboli14) have rarely been described. In a retrospective review10 of disease in elasmobranchs, the prevalence of cardiovascular disease was 5.5%. In that same study, infectious/inflammatory and nutritional etiologies were the most common cause of death in 45% of the cases submitted for histopathology. In a second large-scale retrospective review,15 infectious/inflammatory and nutritional diseases accounted for nearly 74% of all diagnoses. Cardiovascular disease as a cause of death was not identified in the second study; however, cases of vasculitis and myocarditis were noted to have been included under the infectious/inflammatory category. In domestic species and elasmobranchs, infectious, inflammatory, and nutritional etiologies have known associations with cardiac pathology suggesting that the actual prevalence of cardiac disease as a secondary etiology in these species may be higher.11,16,17
According to a 2008 census performed by the American Elasmobranch Society, approximately 9,500 elasmobranchs were reportedly under managed care worldwide. In spite of this, a standardized echocardiographic protocol has not been established for any stingray species at the time of writing. Establishing reference intervals is an essential component of understanding normal cardiac anatomy and physiology as well as pathophysiology in each species. The purpose of this study was to describe an echocardiographic technique for imaging stingrays and to establish associated reference parameters for southern stingrays (Hypanus americanus). Animals enrolled in the study included wild, semiwild, and aquarium-housed southern stingrays. We hypothesized that echocardiographic measurements would not be different between these housing environments. Additionally, we hypothesized that there would be differences in echocardiographic parameters between animals of different size, sex, restraint method (anesthesia or manual restraint), and echocardiographic approach (ventral versus dorsal).
Materials and Methods
Animals
Study animals were 91 subadult to adult southern stingrays from 4 different environments: 2 groups that were under managed care in 3 aquariums (aquarium housed), 1 group that was under managed care in an ocean lagoon at Castaway Cay in The Bahamas (semiwild), and 1 group of wild animals (wild) sampled in Bimini, The Bahamas during wildlife health assessments. Subadult animals were all female and determined to be nulliparous based on the size of the ovary, trophonema, and uterus evaluated during a reproductive ultrasound performed at the time of echocardiography. All animals were defined as healthy based on the results of clinical examination (physical exam, coelomic ultrasound, and in some cases, coelomic fluid analysis) and the absence of any clinicopathological derangements based on concurrent hematology and plasma or serum biochemistry analyses. A total of 7 animals were excluded based on the following: (1) animals were too large to move into dorsal recumbency (1 wild female, 1 semiwild female), (2) animals were too gravid for safe restraint and handling (2 semiwild females), or (3) animals had improper images collected or incomplete studies (1 aquarium-housed female, 2 semiwild females). This resulted in 84 animals being included in the study. Disc width measurements (cm) were collected ventrally by measuring the span between the tips of the widest portion of the pectoral fins. Animal care and use in this project were approved by the animal care and welfare committee at Disney’s Animals, Science and Environment, project No. IR 1806. The Department of Fisheries of the Commonwealth of The Bahamas granted a permit to conduct scientific research (MAF/LIA/22).
Housing and management
Aquarium-housed stingrays were kept at three facilities: the Seas with Nemo and Friends at Epcot (n = 6 females) and Typhoon Lagoon (4 males), which are both parts of Walt Disney World Parks & Resorts, and the Mississippi Aquarium (2 males). The animals at Epcot were housed in a closed, 21.6-million-liter indoor synthetic salt system of mixed species that mimics the natural photoperiod of Florida. The animals at Typhoon Lagoon were housed in a 1.4-million-liter outdoor, closed, synthetic salt system that was strictly regulated to account for rainwater. Animals from the Mississippi Aquarium were housed in a 1.1-million-liter indoor synthetic saltwater aquarium of mixed species and a natural photoperiod that was nearly identical to that of the Walt Disney World Parks & Resorts. Stingrays were fed a variety of fish and shellfish (capelin, clam, shrimp, herring, and smelt) and were supplemented with elasmobranch vitamins (Mazuri Vita-Zu Shark/Ray; Mazuri Exotic Animal Nutrition) 1 to 3 times weekly. Housing at all facilities followed established routine husbandry and water quality protocols used in aquaria.13
Semiwild stingrays (n = 44 females) were housed in an enclosed ocean lagoon under natural environmental conditions at Castaway Cay, The Bahamas, but they were fed a controlled daily diet consisting of shrimp, squid, and a gel product (Aquatic gel 57W9; Mazuri Exotic Animal Nutrition), although animals also had access to natural prey items.
Wild stingrays (n = 1 male, 27 females) were collected and released in the shallow mangrove-fringed waters near South Bimini, The Bahamas.
All animals from each location had full physical examinations performed, which included morphometrics, cardiac and coelomic ultrasound, as well as blood collection for hematology and plasma or serum biochemistry analysis. The aquarium-housed and semiwild animals had these assessments performed annually.
Restraint
Stingrays were examined under manual restraint or anesthesia. Managed animals were routinely examined under general anesthesia using 55 to 75 ppm buffered tricaine methanesulfonate (MS-222; Syndel). Wild animals were obtained by encircling them with a long seine net deployed from a boat. From the confines of the seine net, each animal was gently netted and moved to a 568-liter holding tank on the boat or on shore. All animals were restrained for the echocardiogram by gently holding the pectoral fins. The occasional fluttering of the wings and curling of the pelvic fins were noted during the 3- to 5-minute echocardiogram. After the procedure, animals were released near the shore around the mangroves where they were captured. Given the status of southern rays as a prey species in this area, wild animals were not anesthetized so that they could be released immediately following sampling. All animals in the study were tagged with identification transponders, which ensured that no animal was scanned twice.
Echocardiography
Echocardiography was performed by 2 aquatic animal clinicians (NDM and AMD), and all measurements were performed by a registered sonographer (RTB) under the guidance of a boarded cardiologist (TJG). Ultrasound images were collected using one of the following machines depending on the location; the associated transducer was selected based on optimization of the image quality: a Sonosite 180 Plus (Sonosite, Inc) with either a C60 (2 to 5 MHz) curved array or L52 (5 to 10 MHz) linear transducer, a Sonosite EdgeII C11x (5 to 8 MHz) curved array transducer (Sonosite, Inc), an Ibex Pro (E.I. Medical Imaging) with a CLi3.8 (2.5 to 5 MHz) curved array or L6.2 5- to 8-MHz linear transducer, or a GE Logiqbook with a 4C-RS (2 to 5.5 MHz) or 8C-RS (4 to 10 MHz) curved array transducer (GE Healthcare). For each echocardiographic parameter, measurements were made in triplicate on consecutive beats with the average of the measurements recorded for each animal. The measurements and a subset of images were then reviewed by a cardiologist (TJG) and a sonographer (RTB) to assess the data for validity of establishing echocardiographic reference parameters.
Images depicting the technique for handling and echocardiographic imaging of an anesthetized, aquarium lagoon-housed male Southern stingray in the long-axis (A) and short-axis (B) planes.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0021
Echocardiographic images obtained from southern stingrays depicting standard linear measurements. A—Short-axis view of the ventricle during diastole. B—Short-axis view of the ventricle during systole from the same animal shown in A. Note the prominent ventricular trabeculations in A and B, which were not included in the wall measurements. C—Long-axis view of the sinus venosus, atrium, ventricle, and pericardial sac with fluid. AVD = Atrioventricular valve diameter. VDWd and VDWs = Ventricular dorsal wall during diastole and systole, respectively. VIDd and VIDs, Ventricular internal diameter during diastole and systole, respectively. VL = Ventricular length. VVWd and VVWs = Ventricular ventral wall during diastole and systole, respectively.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0021
With the transducer maintained in a transverse view of the ventricle, the tail of the probe was tilted caudally allowing the image to fan cranially until the sinoatrial valve was observed. The sinoatrial valve diameter (SVD) was measured from the hinge points of the valve leaflets during maximum excursion (Figure 3). The transducer was then moved slightly cranial in the same transverse plane to observe the conus and the conal valve. The conal valve diameter (CVD) was measured with the leaflets at maximal excursion. The probe was then rotated approximately 90° in the clockwise direction to evaluate the conus again in a sagittal plane (Figure 4).
Echocardiographic images obtained from southern stingrays depicting standard linear measurements. A—Modified long-axis view of the heart optimized for visualization of the sinoatrial valve. The sinoatrial valve diameter (SVD) was measured at the hinge points of the valve during maximum excursion. B—Short-axis view of the conal valve at maximum excursion depicting where the conal valve diameter (CVD) was measured. AV = Atrioventricular (valve). CV = Conal valve. P = Pericardial effusion.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0021
Echocardiographic image obtained from a female southern stingray depicting a long-axis image optimized for visualization of the muscular conus arteriosus in its sagittal plane. P = Pericardial effusion.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0021
Measurement of the atrial dimensions was attempted; however, the plasticity of this anatomical structure precluded the ability to make meaningful measurements. Doppler interrogation of flow through the valves was unable to be performed due to an inability to align the ultrasound beam parallel to blood flow given the dorsoventral compression of stingrays. Finally, diastolic and systolic ventricular volume measurements were attempted by tracing the endocardial border from the ventral hinge point of the atrioventricular valve to the dorsal hinge point. Due to the spongy appearance of the endomyocardium, these measurements proved difficult to accurately obtain and were therefore not reported.
A subset of the semiwild animals (n = 25 females) were also restrained in ventral recumbency with the echocardiogram being performed as described above to assess whether or not recumbency had an effect on the measurements. In this subsample of the population, all individuals were evaluated by a single observer (NDM) in both positions, and all measurements were made by the same individual (RTB). All animals were anesthetized with MS-222 while these comparison echocardiograms were performed.
Statistical analysis
Data were analyzed using commercial statistical software with significance set with a P < .05 (SPSS Statistics for Macintosh, Version 27.0; IBM Corp). Distribution of the continuous data (disc width, AVD, VL, CVD, SVD, VVWd, VIDd, VDWd, VVWs, VIDs, VDWs, and FS) was assessed for normality using the Shapiro Wilks test and by examining skewness, kurtosis, and q-q plots. Normally distributed continuous data are reported as mean, standard deviation, and minimum-maximum. Nonnormally distributed continuous data are reported as median and minimum-maximum. A 90% reference interval (RI) was calculated and reported for both nonnormally and normally distributed data. Data were first surveyed for outliers using histograms and normality tests. Outliers were excluded from the dataset until the data did not violate normality. The 90% RI was calculated using robust methods18 with 90% confidence intervals (CI) of reference limits. The 90% CIs were all within < 0.2 of the RI. For size analysis, animals were divided into 2 groups: animals with a disc width of < 80 cm and those with a disc width of > 80 cm. This was determined by evaluating plots of the data, which showed distinct populations. Additionally, 80 cm is approximately when female southern stingrays have been reported to reach sexual maturity.19,20 To evaluate the impact of disc width (< 80 cm and > 80 cm), sex, and the restraint type on the following echocardiographic parameters, an independent t test was performed: AVD, VL, VIDd, VVWs, VDWs, and FS. A Mann-Whitney U test was used to evaluate the impact of disc width, sex, and restraint type on CVD, SVD, VVWd, VDWd, and VIDs. A one-way ANOVA was performed to evaluate differences between echocardiographic measurements based on the animals’ environment. Homogeneity of variance was evaluated for normally distributed data using Levene’s test. The Welch test was used when homogeneity of variance was violated to obtain a corrected F statistic. Post hoc testing to determine differences between groups was performed with the Tukey-Kramer test. Nonnormally distributed data were evaluated by the Kruskal-Wallis test to determine if a difference in distribution across locations was observed. If a difference was observed, then a pairwise comparison was performed to determine the differences between groups.
A subsample (n = 25 females) of the population was evaluated to determine if the dorsal or ventral approach impacted echocardiographic measurements. A paired sample t test was used to evaluate the parametric parameters, while a Wilcoxon test was used to evaluate nonparametric values.
Results
Twenty-eight wild, 44 semiwild, and 12 aquarium-housed southern stingrays (Hypanus americanus) were considered healthy and included in the study. These animals all underwent echocardiographic evaluations for a total of 84 individual exams. There were 9 males and 75 females. Body weight was inconsistently recorded due to the technical difficulty of weighing animals under field conditions. The disc width, a marker of body size independent of weight, was utilized instead and ranged from 46 cm to 111.5 cm. Echocardiography was performed without difficulty, and both short-axis and long-axis views were obtained in all animals allowing for the establishment of reference parameters for markers of cardiac size and function (Table 1). The amount of pericardial effusion appeared similar between all individuals in this study and was subjectively mild (< 0.5 cm).
Descriptive statistics for the total population of southern stingrays evaluated.
| Echocardiographic measurement | Mean | Median | SD | 90% Reference interval | Minimum | Maximum | Normally (N) or nonnormally distributed (NN) |
|---|---|---|---|---|---|---|---|
| Atrioventricular valve diameter (n = 83) | 1.41 | 0.27 | 0.96–1.87 | 0.66 | 2.16 | N | |
| Ventricular length (n = 74) | 3.08 | 0.57 | 2.13–4.05 | 1.71 | 4.53 | N | |
| Conal valve diameter (n = 82) | 1.78 | 1.29–2.22 | 0.89 | 2.21 | NN | ||
| Sinoatrial valve diameter (n = 69) | 0.97 | 0.675–1.17 | 0.59 | 1.79 | NN | ||
| Ventricular ventral wall (diastole) (n = 83) | 0.48 | 0.33–0.59 | 0.25 | 0.90 | NN | ||
| Ventricular dorsal wall (diastole) (n = 83) | 0.50 | 0.31–0.69 | 0.26 | 1.00 | NN | ||
| Ventricular internal diameter (diastole) (n = 83) | 1.59 | 0.30 | 0.966–2.20 | 0.78 | 2.71 | N | |
| Ventricular ventral wall (systole) (n = 83) | 0.64 | 0.15 | 0.37–0.87 | 0.32 | 1.00 | N | |
| Ventricular dorsal wall (systole) (n = 83) | 0.66 | 0.15 | 0.39–0.92 | 0.31 | 1.10 | N | |
| Ventricular internal diameter (systole) (n = 83) | 0.84 | 0.45–1.29 | 0.3 | 1.38 | NN | ||
| FS = [(VIDd − VIDs)/VIDd] X 100 (n = 83) | 44.33 | 9.84 | 26.77–60.67 | 20 | 70 | N |
All measurements are in centimeters with the exception of fractional shortening, which is listed as a percentage. The ventricular length and atrioventricular and sinoatrial valve diameters were all measured in diastole. The conal valve diameter was measured at maximum excursion (ie, during ventricular systole).
VIDd = Ventricular internal diameter during diastole. VIDs = Ventricular internal diameter during systole.
While statistical significance was reached for some of the measurements when comparing restraint and environment, these were not clinically significant. The creation of separate reference intervals between these populations would have greatly depleted the power of the dataset rendering them less useful for clinical practice. Since significant sexual dimorphism occurs in this species, and because linear dimensions (wall thicknesses, valve diameters, and chamber sizes) generally have positive correlations to size, significant differences were expected and observed between both sex and disc width for most of the variables. For these reasons, reference parameters were created for animals based on the disc width using a cutoff of 80 cm (Table 2). While no males were included in the dataset for animals having a disc width > 80 cm, a few smaller females were included in the dataset for animals measuring < 80 cm.
Descriptive statistics for southern stingrays based on size.
| Echocardiographic measurement/disc width | Mean | Median | SD | 90% Reference interval | Minimum | Maximum | Normally (N) or nonnormally distributed (NN) |
|---|---|---|---|---|---|---|---|
| Atrioventricular valve diameter | |||||||
| < 80 cm | 1.23 | 0.28 | 0.70–1.64 | 0.79 | 1.90 | N | |
| > 80 cm | 1.47 | 0.19 | 1.13–1.84 | 0.91 | 1.80 | ||
| Ventricular length | |||||||
| < 80 cm | 2.56 | 0.41 | 1.74–3.21 | 1.78 | 3.51 | N | |
| > 80 cm | 3.33 | 0.46 | 2.57–4.02 | 2.34 | 4.53 | ||
| Conal valve diameter | |||||||
| < 80 cm | 1.4 | 0.97–1.84 | 1.07 | 1.95 | NN | ||
| > 80 cm | 1.9 | 1.50–2.17 | 1.45 | 2.21 | |||
| Sinoatrial valve diameter | |||||||
| < 80 cm | 0.90 | 0.66–1.15 | 0.66 | 1.2 | NN | ||
| > 80 cm | 0.94 | 0.70–1.60 | 1.79 | 1.79 | |||
| Ventricular ventral wall (diastole) | |||||||
| < 80 cm | 0.43 | 0.25–0.75 | 0.26 | 0.80 | NN | ||
| > 80 cm | 0.51 | 0.38–0.70 | 0.40 | 0.90 | |||
| Ventricular dorsal wall (diastole) | |||||||
| < 80 cm | 0.47 | 0.24–0.64 | 0.32 | 0.80 | NN | ||
| > 80 cm | 0.55 | 0.39–0.79 | 0.40 | 1.00 | |||
| Ventricular internal diameter (diastole) | |||||||
| < 80 cm | 1.39 | 0.33 | 0.77–1.89 | 1.0 | 2.14 | N | |
| > 80 cm | 1.62 | 0.30 | 1.2–2.23 | 1.18 | 2.25 | ||
| Ventricular ventral wall (systole) | |||||||
| < 80 cm | 0.52 | 0.12 | 0.32–0.86 | 0.36 | 0.90 | N | |
| > 80 cm | 0.70 | 0.12 | 0.53–0.92 | 0.40 | 1.00 | ||
| Ventricular dorsal wall (systole) | |||||||
| < 80 cm | 0.56 | 0.17 | 0.29–0.80 | 0.34 | 0.90 | N | |
| > 80 cm | 0.72 | 0.14 | 0.55–1.0 | 0.48 | 1.10 | ||
| Ventricular internal diameter (systole) | |||||||
| < 80 cm | 0.70 | 0.38–1.06 | 0.53 | 1.22 | NN | ||
| > 80 cm | 0.84 | 0.56–1.33 | 0.50 | 1.38 | |||
| FS = [(VIDd − VIDs)/VIDd] X 100 | |||||||
| < 80 cm | 45.1 | 8.2 | 31–59 | 32 | 58 | N | |
| > 80 cm | 45.3 | 11.1 | 26–60 | 20 | 70 |
Animals were categorized as either being < 80 cm disc width (n = 22) or > 80 cm disc width (62). All measurements are in centimeters with the exception of fractional shortening, which is listed as a percentage. The ventricular length and atrioventricular and sinoatrial valve diameters were all measured at maximum excursion during ventricular diastole. The conal valve diameter was measured at maximum excursion during ventricular systole.
The AVD (t = −5.158, P < .001), VL (t = −6.69, P < .001), VIDd (t = −3.99, P < .001), VVWs (t = −6.02, P < .001), VDWs (t = −4.69, P < .001), CVD (U = 28.27, P < .001), VVWd (U = 14.810, P < .001), VDWd (U = 12.44, P < .001), and VIDs (U = 11.373, P < .001) were determined to be significantly different based on size with 80 cm as a cutoff. Fractional shortening (t = 0.627, P = .532) and SVD (U = 2.639, P = .104) were not statistically significant between the 2 size groups.
The AVD (t = 3.467, P < .001), VL (t = 3.095, P = .001), VIDd (t = 2.609, P = .011), VVWs (t = 4.025, P < .001), VDWs (t = 3.931, P < .001), CVD (U = 16.067, P < .001), VVWd (U = 11.626, P < .001), VDWd (U = 10.608, P < .001), and VIDs (U = 6.782, P = .009) were determined to be significantly different between sexes (Supplementary Table S1), while FS (t = 0.356, P = .723) and SVD (U = 0.837, P = .360) were not significantly different.
The AVD (t = −0.800, P = .426), SVD (U = 0.597, P = .551), VIDs (U = −1.480, P = .139), VIDd (t = 0.797, P = .214), VDWs (t = −0.871, P = .386), and the FS (t = −0.668, P = .506) were not significantly different between animals restrained manually and those who were anesthetized, while VL (t = −3.165, P = .002), VVWs (t = −3.377, P < .001), CVD (U = 3.536, P < .001), VVWd (U = 4.23, P < .001), and VDWd (U = 2.30, P = .021) were determined to be significantly different (Supplementary Table S1).
The AVD was not significantly different (F = 1.895, P = .157) based on environment, while VL was significantly different (F = 6.63, P = .002). Differences were observed between wild, semiwild (P = .004), and aquarium-housed (P = .010) populations. No difference was observed between the aquarium-housed population and the semiwild population (P = .718). A significant difference was observed in CVD based on environment [H(2), 8.9; P = .012]. The difference was observed between the wild and semiwild populations (P = .036) and between the wild and aquarium-housed populations (P = .025). A significant difference was observed in SVD based on environment [H(2), 17.902; P < .001]. The difference was observed between the wild and semiwild (P = .020) populations and between the semiwild and aquarium-housed populations (P < .001). A significant difference was observed in VVWd based on environment [H(2), 19.686; degrees of freedom (Df), 2; P < .001]. The difference was observed between the wild and semiwild (P < .001) populations and between the wild and aquarium-housed populations (P = .007). A significant difference was observed in VDWd based on the environment [H(2), 10.914; Df, 2; P = .004]. The difference was observed between the wild and semiwild (P = .004) populations and the semiwild and aquarium-housed populations (P = .035). The VVWs were determined to be significant based on the environment (F = 6.23, P = .003). Differences were observed between wild and semiwild populations (P = .002). No significant difference was observed between the environment and VIDd (F = 0.596, P = .554), VIDs (F = 1.78, P = .230), VDWs (F = 2.55, P = .084), or FS (F = 0.333, P = .718). These data are reported elsewhere (Supplementary Table S2).
Measurements of AVD (t = −0.223, P = .825), VL (t = 0.775, P = .446), CVD (t = 0.876, P = .389), SVD (W = −0.975, P = .330), VIDd (t = 0.459, P = .650), VDWd (t = 0.227, P = .822), VIDs (t = −0.571, P = .573), and FS (W = 0.038, P = .970) were not statistically different based on the echocardiographic approach, while VVWd (t = 2.77, P = .010, VVWs (W = −2.44, P = .015), and VDWs (t = 2.93, P = .007) were statistically different depending on the approach (Supplementary Table S1).
Discussion
This study provides echocardiographic reference parameters for clinically healthy southern stingrays (Hypanus americanus). Echocardiography remains the most widely utilized tool for evaluating the cardiac size and function in most species, it is more readily available to zoo and aquatic animal veterinarians than other imaging modalities, and it is feasible in stingrays. The echocardiographic parameters established in this study can be used to assess cardiac health in this species. While flow velocities could not be obtained due to anatomical constraints, all chambers and valves were consistently visualized, and standard 2-dimensional measurements (eg, wall thicknesses and ventricular chamber sizes) were obtained in all animals. Evaluation of the conal and atrioventricular valve morphology was consistently achieved, whereas the sinoatrial valve was identified less frequently. Indeed, the sinoatrial valve was still consistently visualized in a majority of the animals studied. The inconsistent visualization of this valve was suspected to be due to the thin-walled, relatively mobile nature of the sinus venosus and atrium both connected by the sinoatrial valve. No significant valvular regurgitation was noted in any animal in this study. Diastolic and systolic ventricular volume measurements were attempted by using Simpson’s method of discs, which involves tracing the endomyocardial border from the hinge points of the atrioventricular valve as has been done in other species.21 These measurements are used to calculate ejection fraction, a marker of systolic function. Because the myocardium of stingrays has a spongy appearance at both the macroscopic and microscopic levels,22,23 attempts were made to exclude the trabeculations from the measurements. Instead, the compact myocardial borders were used for tracing. This proved to be difficult, and the measurements were not consistently obtained among animals and were excluded from the analysis. Given the marked variation in ventricular morphology between elasmobranch species,24,25 these measurements may prove viable in other species and should be attempted.
While clinically relevant statistical differences were expected and observed based on size and between sexes given the marked sexual dimorphism in this species, significant differences were also noted between animals housed in different environments, between the restraint technique, and between the echocardiographic approaches. While most of these differences were clinically insignificant and presumably related to the relatively small individual datasets, a brief review of elasmobranch cardiac physiology is vital to understanding how some echocardiographic measurements may be affected by the housing environment, handling, and approach within and between stingray species.
Cardiac output is defined as the product of heart rate and stroke volume. Cardiac physiology in elasmobranchs is unique in that these species rely much more heavily on stroke volume than on alteration of the heart rate to alter cardiac output. Heart rate remains relatively constant between periods of exercise and rest in elasmobranchs.24–26 Whereas alterations in stroke volume are mostly driven by changes in vascular and autonomic tone in mammals, elasmobranchs rely heavily on changes in pericardial pressure to regulate their stroke volume.24–26 Elasmobranchs have a semirigid, noncompliant pericardium that holds a variable amount of fluid.27–29 Given the rigidity, more negative pericardial pressures can be generated during ventricular systole as compared to the more compliant pericardium of mammals. A more negative pericardial pressure leads to an increase in the transmural (distending) pressure of both the atrial and ventricular walls, which increases venous return. This increased venous return enhances the Frank-Starling mechanism leading to increased stroke volumes. Several studies27–29 have demonstrated the ability of elasmobranchs to alter the volume of pericardial fluid by way of the pericardioperitoneal canal. During high output states (eg, feeding, disease, exercise), pericardial fluid is evacuated from the pericardial space through this canal allowing for a decrease in the pericardial pressure. Since pericardial pressure and cardiac output are inversely related, this results in an increase in the stroke volume thereby augmenting cardiac output.
The ability to rapidly alter stroke volume by way of the pericardioperitoneal canal may have a pronounced effect on echocardiographic measurements in elasmobranchs. Cardiac filling has a direct effect on both chamber dimensions and wall thicknesses, with cavitary volume having an inverse relationship with wall thickness. For example, reduced left ventricular chamber dimensions and thickened left ventricular walls are sometimes observed in animals afflicted with dehydration and/or hypovolemia and are referred to as pseudohypertrophy.30 Conversely, hemodynamic states resulting in increased circulating blood volumes such as anemia and certain metabolic diseases can increase chamber size and cause relative thinning of the ventricular walls.31,32 Abel et al27 measured pericardial fluid volumes in horn sharks and found that 22% percent of the pericardial fluid volume was ejected immediately following handling. In that same study, 1 animal ejected 78% of its pericardial fluid resulting in a dramatic increase in the subsequent pericardial pulse pressure, an indirect marker of stroke volume. Based on the relationship between stress, pericardial fluid volume, and cardiac filling, animals experiencing a higher level of stress during handling would be expected to have larger ventricular chamber sizes (diameters and lengths), smaller ventricular wall measurements, larger valve diameters (due to annular stretch), and higher fractional shortening and ejection fractions when compared to nonstressed animals.
Some of the statistically significant differences noted between the populations (environment, restraint, and echocardiographic approach) in this study may have been the result of variations in the pericardial fluid volume. It was initially hypothesized that wild animals and those who underwent manual restraint would have higher levels of stress as compared to the aquarium-housed animals. As part of a simultaneous study using a fraction of this dataset, hematogenous 1α-hydroxycorticosterone levels, a unique corticosteroid in elasmobranchs, were measured between animals from the three different environments as a marker to compare relative stress. All of the animals showed low levels of 1α-hydroxycorticosterone regardless of location, handling technique, or sex.33 Notably, VIDs, VIDd, and FS, markers of ventricular chamber size and function, were not statistically different between housing environments nor restraint type in this study. This suggests that pericardial fluid volume was not augmented enough to affect cardiac output and likely had little effect, if any, on the other cardiac measurements in this study.
The statistically significant differences observed between animals within different environments (wild, semiwild, and aquarium-housed) may be relatable to variations in diet, fitness, and parasite load. Wild animals would be expected to have a more variable diet and the potential for improved fitness as compared to aquarium-housed animals. The semiwild animals likely experience conditions somewhere between the other 2 groups. Additionally, parasitic diseases, especially those with the propensity to cause fluctuations in blood and/or red blood cell volumes, may secondarily affect the filling of the cardiac chambers as previously discussed. Wild and semiwild animals may have had higher parasite loads when compared to aquarium-housed animals receiving regular veterinary care. Similarly, anesthesia can affect ventricular filling and inotropy depending on the anesthetic agents used, so the statistically significant differences between the anesthetized and manually restrained animals were not an unexpected finding. Further study comparing size and sex-matched animals from the same environment would be necessary to directly evaluate the significance of anesthesia alone on echocardiographic measurements.
When comparing variables obtained between dorsal and ventral approaches, only VVWd, VVWs, and VDWs were significantly different. Again, these differences were very small, unlikely to be clinically relevant, and might have occurred due to the combined effect of differences in the compressive force of the transducer on the body wall and the position of the heart within the pericardial fluid. Based on these findings, the authors believe either approach to be acceptable for echocardiography in this species.
Of the statistically significant differences observed between environment, sex, size, restraint, and echocardiographic approach for some of the variables, only size and sex were deemed clinically relevant. Given the pronounced sexual dimorphism of this species, the significant differences documented between the sexes were likely more dependent on size than true differences between the sexes. For this reason, echocardiographic parameters were developed based on the disc width, which effectively separated most males and females. Using a cutoff of 80 cm for disc width, significant differences were noted between the 2 populations for all variables except FS, which is generally a weight-independent parameter in most species.
This study had limitations, which included the aforementioned differences related to housing environment and restraint and the relatively small sample size. While it is prudent to establish echocardiographic reference parameters in a uniform population free from variations in the handling technique and housing environment, the authors felt the utility of a larger dataset outweighed the desire to establish parameters in a much smaller uniform population. For this reason, the dataset was combined for analysis using a size cutoff instead of one based on environment, sex, or handling technique. Robust statistical comparisons were performed in an effort to disclose statistical differences between the populations. Future studies may choose to elaborate on these established parameters within a more uniform population. Had more animals been included for statistical comparison between the aforementioned variables, some of the statistically significant differences may have been resolved. Further, while the echocardiogram was performed using the same technique between all populations, field conditions inevitably create the potential to affect subtle differences in image acquisition.
In conclusion, echocardiography was feasible in southern stingrays, and echocardiographic parameters were established. Statistical significance was reached when comparing measurements from animals in different environments, between handling techniques, and between echocardiographic approaches, but these were not considered clinically relevant. Statistically and clinically significant differences were noted based on sex and size, which led to the establishment of echocardiographic reference parameters separated based on size (disc width). This effectively separated most males and females. Since there is evidence noting differences in cardiac structure and function between stingray species attributable to size and lifestyle (eg, benthic vs pelagic species), the reference parameters established in this study should not be extrapolated to other species.22,23,34
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
This study was supported in part by the Bahamas Department of Marine Resources, the National Institute of General Medical Sciences (NIGMS) grant No. P20GM104932, and COBRE, CORE-NPN, Chemistry Research Core. Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript.
The authors have nothing to declare.
The authors thank Jayur Patel for his anatomy sketches that aided in the development of the echocardiographic technique and measurements, the volunteers and field assistants at the Bimini Biological Field Station Foundation, and the animal health and husbandry staff at Disney’s Castaway Cay; Disney’s Animals, Science and Environment; the Georgia Aquarium; and the Mississippi Aquarium for their assistance during the examinations. We also thank the Great Ape Heart Project for allowing the use of their database for data measurement and analysis.
References
- 1.↑
Compagno LJV. Checklist of living elasmobranchs. In: Hamlett WC, ed. Sharks, Skates, and Rays: the Biology of Elasmobranch Fishes. John Hopkins University Press; 1999:471–498.
- 2.↑
Borucinska JD, Obasa OA, Haffey NM, et al. Morphological features of coronary arteries and lesions in hearts from five species of sharks collected from the northwestern Atlantic Ocean. J Fish Dis. 2012;35(10):741–754. doi:10.1111/j.1365-2761.2012.01405.x
- 3.
Cox GK, Brill RW, Bonaro KA, et al. Determinants of coronary blood flow in sandbar sharks (Carcharhinus plumbeus). J Comp Physiol. 2017;187(2):315–327. doi:10.1007/s00360-016-1033-x
- 4.
Satchell GH, Johnes MP. The function of the conus arteriosus in the Port Jackson shark, Heterodontus portusjacksoni. J Exp Biol. 1967;46(2):373–382. doi:10.1242/jeb.46.2.373
- 5.↑
Tebecis AK. A study of electrograms recorded from the conus arteriosus of an elasmobranch heart. Aust J Biol Sci. 1967;20(4):843–846. doi:10.1071/BI9670843
- 6.↑
Hyatt M, Gerlach TJ. Diagnosis and management of suspected congestive heart failure secondary to dilated cardiomyopathy in a sand tiger shark (Carcharius taurus) with establishment of preliminary normal echocardiographic indices. J Zoo Wildl Med. 2022;53(2):363–372.
- 7.↑
Vergneau-Grosset C, Summa N, Rapoport G, et al. Management of a dilated cardiomyopathy in a leopard shark (Triakis semifasciata). In: Proceedings of the 51st International Association for Aquatic Animal Medicine Forum. IAAAM; 2020:virtual.
- 8.↑
Adamson ML, Caira JN. Lockenloia sanguinis n. gen., n. sp. (Nematoda: Dracunculoidea) from the heart of a nurse shark, Ginglymostoma cirratum, in Florida. J Parasitol. 1991;77(5):663–665. doi:10.2307/3282695
- 9.↑
Caira JN, Benz GW, Borucinska J, Kohler NE. Pugnose eels, Simenchelys parasiticus (Synaphobranchidae) from the heart of a shortfin mako, Isurus oxyrinchus (Lamnidae). Environ. Biol. Fishes. 1997;49(1):139–144. doi:10.1023/A:1007398609346
- 10.↑
Garner MM. A retrospective study of disease in elasmobranchs. Vet Pathol. 2013;50(3):377–389. doi:10.1177/0300985813482147
- 11.↑
Perpiñán D, Costa T. Metastatic mineralization in blacktip reef sharks (Carcharhinus melanopterus). J Fish Dis. 2017;40:447–451. doi:10.1111/jfd.12524
- 12.↑
Mylniczenko ND, Kinsel M, Young F, et al. Lessons from a case of capture myopathy in a pelagic silky shark (Carcharhinus falciformis). In: Proceedings of the 35th Annual American Association of Zoo Veterinarians Forum. AAZV; 2003:211–216.
- 13.↑
Mohan PJ, Aiken AN. Water quality and life support systems for large elasmobranch exhibits. In: Smith M, Warmolts D, Thoney D, Hueter R, eds. The Elasmobranch Husbandry Manual: Captive Care of Sharks, Rays and Their Relatives. Ohio Biological Survey; 2004:69–88.
- 14.↑
Raymond JT, Dunker F, Garner MM. Ink embolism in freshwater orange spot stingrays (Potamotrygon motoro) following tattooing procedure. In: Proceedings of the 35th Annual American Association of Zoo Veterinarians Forum. AAZV; 2003:210.
- 15.↑
Stidworthy MF, Thornton SM, James R. A review of pathologic findings in elasmobranchs: a retrospective case series. In: Smith M, Warmolts D, Thoney D, Hueter R, eds. The Elasmobranch Husbandry Manual II: Recent Advances in the Care of Sharks, Rays and Their Relatives. Ohio Biological Survey; 2017:277–286.
- 16.↑
Freeman L, Rush J, Adin D, et al. Prospective study of dilated cardiomyopathy in dogs eating nontraditional or traditional diets and in dogs with subclinical cardiac abnormalities. J Vet Intern Med. 2022;36:451–463. doi:10.1111/jvim.16397
- 17.↑
Lakhdhir S, Viall A, Alloway E, Keene B, Baumgartner K, Ward J. Clinical presentation, cardiovascular findings, etiology, and outcome of myocarditis in dogs: 64 cases with presumptive antemortem diagnosis (26 confirmed postmortem) and 137 cases with postmortem diagnosis only (2004-2017). J Vet Cardiol. 2020;30:44–56. doi:10.1016/j.jvc.2020.05.003
- 19.↑
Henningsen AD, Leaf RT. Observations on the captive biology of the southern stingray. Trans Am Fish Soc. 2010;139(3):783–791. doi:10.1577/T09-124.1
- 20.↑
Ramirez-Mosqueda E, Perez-Jimenez JC, Mendoza-Carranza M. Reproductive parameters of the southern stingray in southern Gulf of Mexico. Lat Am J Aquat. 2012;40(2):335–344. doi:10.3856/vol40-issue2-fulltext-8
- 21.↑
Boon JA. Appendix IV: canine. In: Veterinary Echocardiography. 2nd ed. Wiley-Blackwell; 2011:678–703.
- 22.↑
Tota B. Myoarchitecture and vascularization of the elasmobranch heart ventricle. J Exp Zool. 1989;252(S2):122–135. doi:10.1002/jez.1402520413
- 23.↑
Tota B, Gattuso A. Heart ventricle pumps in teleosts and elasmobranchs: a morphodynamic approach. J Exp Zool Part A Ecol Genet Physiol. 1996;275:162–171. doi:10.1002/(SICI)1097-010X(19960601/15)275:2/3<162::AID-JEZ8>3.0.CO;2-B
- 24.↑
Brill RW, Lai NC. Elasmobranch cardiovascular system. In: Shadwick RE, Farrell AP, Brauner C, eds. Fish Physiology: Physiology of Elasmobranch Fishes, Internal Processes. Elsevier; 2015:1–82.
- 25.↑
Butler PJ, Metcalfe JD. Cardiovascular and respiratory systems. In: Shuttleworth TJ, ed. Physiology of Elasmobranch Fishes. Springer-Verlag; 1988:1–47.
- 26.↑
Lai NC, Graham JB, Lowell WR, et al. Elevated pericardial pressure and cardiac output in the leopard shark (Triakis semifasciata) during exercise: the role of the pericardioperitoneal canal. J Exp Biol. 1989;147(1):263–277. doi:10.1242/jeb.147.1.263
- 27.↑
Abel DC, Graham JB, Lowell WR, Shabetai R. Elasmobranch pericardial function. 1. Pericardial pressures are not always negative. Fish Physiol Biochem. 1986;1(2):75–83. doi:10.1007/BF02290207
- 28.
Abel DC, Lowell WR, Graham JB, Shabetai R. Elasmobranch pericardial function. 2. The influence of pericardial pressure on cardiac stroke volume in horn sharks and blue sharks. Fish Physiol Biochem. 1987;4(1):5–14. doi:10.1007/BF02073861
- 29.↑
Abel DC, Lowell WR, Lipke MA. Elasmobranch pericardial function. 3. The pericardioperitoneal canal in the horn shark (Heterodontus francisci). Fish Physiol Biochem 1994;13(3):263–274. doi:10.1007/BF00004364
- 30.↑
Fine DM, Durham HE, Rossi NF, Spier AW, Selting K, Rubin LJ. Echocardiographic assessment of hemodynamic changes produced by two methods of inducing fluid deficit in dogs. J Vet Intern Med. 2010;24(2):348–353. doi:10.1111/j.1939-1676.2009.0448.x
- 31.↑
Cho IJ, Mun YC, Kwon KH, Shin GJ. Effect of anemia correction on left ventricular structure and filling pressure in anemic patients without overt heart disease. Korean J Intern Med. 2014;29(4):445–453. doi:10.3904/kjim.2014.29.4.445
- 32.↑
Mohrman DE, Heller LJ. Cardiovascular physiology. In: Weitz M, Kearns B, eds. Cardiovascular Physiology. 8th ed. McGraw-Hill Education; 2014:193–215.
- 33.↑
Wheaton CJ, Burns CM, Smukall MJ, et al. Measurement of 1α-hydroxycorticosterone: Correlations with secondary stress markers in aquarium-managed, semi-wild, and wild populations of rays and carcharhinid sharks. In: Proceedings of the 35th American Elasmobranch Society Forum. AES; 2020:virtual.
- 34.↑
Grim JM, Ding AA, Bennett WA. Differences in activity level between cownose rays (Rhinoptera bonasus) and Atlantic stingrays (Dasyatis sabina) are related to differences in heart mass, hemoglobin concentration, and gill surface area. Fish Physiol Biochem. 2012;38(5):1409–1417. doi:10.1007/s10695-012-9628-y
