Primary IMHA in dogs is characterized by RBC destruction and is associated with a high mortality rate. A prothrombotic state and subsequent thromboembolic disease is thought to be a major contributor to death of dogs with this disease.1–3 Most dogs with idiopathic IMHA are in a hypercoagulable state at the time of diagnosis, and it is likely that hemolysis is an important instigating factor.2 Several mechanisms have been proposed to cause a hypercoagulable state in IMHA, with the most likely sources being a combination of alterations in cytokine release, number of activated platelets, and concentrations of tissue factor, antithrombin III, fibrinogen, soluble fibrin, and D-dimers.1,2
According to Poiseuille's law, resistance to blood flow is proportional to vessel length and blood viscosity. The decrease in blood viscosity with hemolytic anemia is expected to contribute to a decreased resistance in flow. Experimentally induced normovolemic anemia results in significant increases in PV blood flow in dogs,4 which is similar to results of a study5 performed with an anemic newborn lamb. It is postulated that the complex interactions of an altered hemodynamic state, activated endothelium, and reduced viscosity in dogs with IMHA affect blood flow velocities in various unknown manners. Investigators of an in vitro study6 determined that as blood flow velocity near an endothelial surface increases, shear stress also increases, which results in an increasingly fibrinolytic state.
Standard diagnostic evaluations to determine the primary causes of IMHA include a complete abdominal ultrasonographic examination. Given that most cases of IMHA are primary, results of the abdominal ultrasonographic examination rarely alter the treatment plan or provide additional prognostic information; however, it can be used to exclude secondary causes such as neoplastic disease or severe infection. Pulsed-wave Doppler imaging has been used primarily to evaluate the PV of dogs during the assessment of portosystemic shunts and portal hypertension; mean blood flow velocities of the PV have been reported.7 To our knowledge, quantitative assessment of blood flow velocities in the CVC of the cranial portion of the abdomen in clinically normal or diseased dogs has not been reported; however, there have been qualitative descriptions (a multiphasic waveform as a result of pressure changes in the right atrium), similar to descriptions reported for the hepatic veins of anesthetized dogs.8–12 Measurement of blood flow velocities in these main abdominal veins by use of pulsed-wave Doppler ultrasonography might expand the clinical use of abdominal ultrasonography for differentiating hypercoagulable dogs from clinically normal dogs.
The purpose of the study reported here was to compare peak blood flow velocity in the PV and CVC of clinically normal dogs and dogs with IMHA. We also intended to provide initial reference values for CVC flow velocities in healthy dogs. We hypothesized that clinically affected dogs will have a higher peak blood flow velocity in the PV and CVC, compared with peak blood flow velocity in clinically normal dogs.
Materials and Methods
Animals
Client-owned dogs admitted to the Michigan State University Veterinary Medical Center with suspected primary IMHA that had been newly diagnosed were eligible for inclusion. A diagnosis of IMHA was made on the basis of anemia (Hct ≤ 30%) and presence of ≥ 2 of the following 4 criteria: regenerative anemia, hyperbilirubinemia, spherocytosis, or autoagglutination. Dogs were enrolled during an 8-month period from 2015 to 2016; dogs could participate, provided they had not received anticoagulant treatment. Previous or concurrent treatment with immunosuppressive drugs was allowed. If dogs required administration of blood products at the time of admission, blood samples were obtained before initiation of treatment, and ultrasonographic images were obtained before or at the time of blood product infusion. For all dogs, a complete medical history was obtained, and physical examination, a CBC, serum biochemical analysis, full coagulation profile, TEG, survey thoracic radiography, and abdominal ultrasonography were performed. The full coagulation profile included assessment of prothrombin time, activated partial thromboplastin time, fibrinogen degradation time, and concentrations of fibrinogen, D-dimers, and antithrombin III. All thoracic radiographs and abdominal ultrasonograms were reviewed by a board-certified veterinary radiologist (NCN). Dogs with secondary IMHA or concurrent severe thrombocytopenia were excluded.
Clinically normal dogs owned by hospital staff and veterinary students served as a control group. Adult dogs ≥ 1 year old with body weight ≤ 25 kg were recruited to match the affected patient population. These dogs were deemed healthy on the basis of the medical history and results of physical examination, a CBC, serum biochemical analysis, full coagulation profile, TEG, and abdominal ultrasonography.
Written consent was obtained from all owners for use of all dogs. The study protocol was approved by the Michigan State University Institutional Animal Care and Use Committee.
TEG
Blood samples for TEG were obtained from all dogs via saphenous venipuncture with a 21-gauge butterfly needle; samples were collected into a 3-mL plastic syringe containing 0.3 mL of 3.2% trisodium citrate. All syringes were filled to a total volume of 3 mL (0.3 mL of citrate and 2.7 mL of blood), which yielded a final ratio of 1 part citrate to 9 parts blood. Syringes were inverted gently to ensure adequate mixing of the citrate and blood and then allowed to remain undisturbed at room temperature (23°C) for 45 minutes before analysis.13
Nonactivated TEG was performed by 1 investigator (RPS), who used an automated hemostasis analyzera in accordance with a previously described protocol.13 Quality control samples were assayed immediately before each test, and all samples were assayed in duplicate. Hypercoagulability was defined as an MA value above that for the reference standard of our facility.13
Ultrasonography
One investigator (RPS) performed all ultrasonographic examinations by use of a variable-frequency 5- to 8-MHz convex-array transducer.b Complete abdominal ultrasonography was performed, which was followed by a pulsed-wave Doppler ultrasonographic examination. All B-mode images were obtained at a frequency of 8 MHz. Dogs were not sedated or anesthetized for the examinations. When possible, food was withheld for 12 hours before examination.
The ventral and right lateral aspects of the abdomen were clipped, cleaned, and covered with acoustic gel. Doppler ultrasonographic examinations of the CVC and PV were performed with patients positioned in dorsal recumbency to allow the best visualization of the CVC and PV in a long-axis view; insonation angle was ≤ 60° by use of angle correction, as previously reported for the PV.5 A sample gate of 2 mm was used, with the gate centered in a vessel. All pulsed-wave tracings were obtained simultaneously with an ECG recording, and instantaneous heart rate was computed from the preceding R-R interval on the ECG.
Images of the CVC were obtained at the entrance of the vessel into the liver cranial to the right renal vein via a right ventral intercostal approach, with evidence of the phasic components that correlated to the pressure conditions of the right atrium, as described for a previous study9 of the hepatic veins. All blood flow velocities of the CVC were obtained at end expiration and maximal S-wave velocity (Figure 1). Images of the PV were obtained with an oblique sagittal plane at the level of the right portal branch. All blood flow velocities of the PV were obtained at end expiration and peak of the R wave on the ECG complex (Figure 2).
B-mode ultrasonographic images (A and C) and pulsed-wave Doppler ultrasonographic images (B and D) of the CVC at the porta hepatis in a healthy unsedated dog (A and B) and an unsedated dog affected with IMHA (C and D). In the B-mode images, notice the location (short parallel bars that intersect the long lines) that was used for the placement of the probe for Doppler ultrasonography. In the pulsed-wave Doppler ultrasonographic images, notice the simultaneously recorded ECG tracing. A caliper was placed on the peak of the S wave (+1) to indicate peak velocity toward the heart at end expiration in both dogs.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
B-mode ultrasonographic images (A and C) and pulsed-wave Doppler ultrasonographic images (B and D) of the main PV at the porta hepatis in a healthy unsedated dog (A and B) and an unsedated dog affected with IMHA (C and D). In the pulsed-wave Doppler ultrasonographic images, notice the simultaneously recorded ECG tracing. A caliper was placed on the peak of the R wave (+1) at end expiration in both dogs. There was similar blood flow velocity for both dogs.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
Mean velocity measurements were obtained by calculating the mean value for 3 measurements. Measurements of maximum vessel diameter were obtained by cine looping an image and obtaining the largest value in the transverse plane.
Data analysis
Statistical analyses were performed by a statistician (DJS) by use of commercially available software.c,d Descriptive statistics that included mean, median, SD, and variance were generated for the hemostatic and ultrasonographic measurements with commercially available software.e Distributions of data were tested for normality by use of the Shapiro-Wilk test and examination of the P-P plot. To compare data for the IMHA and control groups, a t test was used when the distributions were normal; the Wilcoxon rank sum test was used when distributions were not normal. A logistic regression model was used to predict the presence of IMHA on the basis of blood flow velocity in the CVC. Values were considered significant at P ≥ 0.05.
Results
Animals
Eleven dogs (7 spayed females and 4 castrated males) were identified as eligible for inclusion in the IMHA group. The IMHA group consisted of 1 each of Miniature Schnauzer, Cocker Spaniel, Dachshund, Toy Poodle, Bichon Frise, Cardigan Welsh Corgi, and Staffordshire Terrier; there were 4 mixed-breed dogs. Median age of dogs with IMHA was 7 years (range, 1.5 to 12 years), and median body weight was 12.2 kg. Seven dogs with IMHA received packed RBC transfusions that were initiated after blood sample collection and during or after ultrasonographic examination. The remaining dogs did not receive packed RBC transfusions during their hospital stay.
Twenty-one clinically normal dogs (15 spayed females, 5 castrated males, and 1 sexually intact female) were enrolled in the control group. The control group consisted of 2 Shetland Sheepdogs, 2 Jack Russell Terriers, and 1 each of Beagle, Border Collie, Pug, Cairn Terrier, Maltese, Boston Terrier, Chihuahua, Pembroke Welsh Corgi, Papillon, Pomeranian, and Staffordshire Terrier; there were 6 mixed-breed dogs. Median age of the control dogs was 4 years (range, 1 to 13 years), and median body weight was 11 kg.
Median heart rate of dogs with IMHA was 117 beats/min (range, 84 to 140 beats/min), whereas median heart rate of the control dogs was 112 beats/min (range, 67 to 185 beats/min). There was no significant difference between groups with regard to age (P = 0.21), body weight (P = 0.06), or heart rate (P = 0.75).
Hemodynamic measurements
Results of hematologic and TEG findings were summarized (Table 1). Dogs with IMHA had a significantly (P = 0.01) lower packed RBC volume, significantly (P = 0.01) higher bilirubin and fibrinogen concentration, and significantly (P = 0.03) longer prothrombin time, compared with results for the control group. Although there was a significant difference in prothrombin time, the median prothrombin value for both groups was within reference limits.
Median (range) values of select clinical variables for 11 dogs with IMHA and 21 clinically normal control dogs.
| Variable | IMHA group | Control group | P value* |
|---|---|---|---|
| Hematologic and biochemical | |||
| PCV (%) | 18 (9–27) | 46 (32–58) | 0.01 |
| Bilirubin (mg/dL) | 2.3 (0.6–18.3) | 0.2 (0.1–0.3) | 0.01 |
| aPTT (s) | 24 (17–153) | 16 (13–24) | 0.09 |
| PT (s) | 8 (7–15) | 7 (7–9) | 0.03 |
| Fibrinogen (mg/dL) | 614 (310–946) | 176 (130–253) | 0.01 |
| TEG | |||
| MA† (mm) | 73.7 (39.9–91.7) | 51.8 (30.5–75.4) | 0.01 |
| R‡ (s) | 11 (0–14) | 21 (6–32) | 0.01 |
| K§ (s) | 3 (1–7) | 11 (2–22) | 0.01 |
| α angle ‖ (°) | 54 (24–88) | 19 (10–68) | 0.01 |
Values were considered significant at P ≤ 0.05.
Represents clot strength.
Represents interval from initiation until initial detection of clot formation.
Represents interval until clot formation.
Represents rate of clot formation.
aPTT = Activated partial thromboplastin time. PT = Prothrombin time.
The TEG variables were consistent with those reported in other studies2,3 of dogs with IMHA. Dogs with IMHA were characterized by significant (P = 0.01) increases in MA and rate of clot formation and significant (P = 0.01) decreases in the interval from initiation to detection of clot formation and the interval to clot formation.
Ultrasonography
Mean ± SD blood flow velocity in the CVC differed significantly (P = 0.01) between the control (63.0 ± 18.6 cm/s) and IMHA (104.0 ± 36.9 cm/s) groups (Figure 3). Median blood flow velocity in the CVC of the control group was 66.6 cm/s (range, 37.5 to 109.8 cm/s), whereas for dogs with IMHA it was 105 cm/s (range, 49.5 to 178.2 cm/s). Variance in blood flow velocity in the CVC of dogs with IMHA (1,362 cm2/s2) was significantly (P = 0.01) greater than that of the control group (363 cm2/s2).
Box-and-whisker plots of peak blood flow velocity in the CVC (with blood flow directed toward the heart) of 11 dogs with IMHA and 21 clinically normal control dogs. Each box represents the 25th through 75th percentiles, the horizontal line in each box represents the median, and the whiskers represent the minimum and maximum values without outliers. *Peak velocity differed significantly (P = 0.01) from the value for the IMHA group.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
No significant difference was detected between IMHA and control groups for mean blood flow velocity in the PV (P = 0.27) or PV velocity variance (P = 0.18; Figure 4). Median blood flow velocity in the PV of the control dogs was 26.7 cm/s (range, 17.4 to 42.4 cm/s), whereas it was 34.9 cm/s (range, 16 to 51.0 cm/s) for the dogs with IMHA. There was also no significant difference between groups with regard to the mean diameter of the CVC (P = 0.07) or PV (P = 0.30). Mean ± SD CVC diameter of the control and IMHA groups was 0.74 ± 0.24 cm and 0.80 ± 0.44 cm, respectively. Mean PV diameter of the control and IMHA groups was 0.49 ± 0.17 cm and 0.59 ± 0.23 cm, respectively.
Box-and-whisker plots of peak blood flow velocity in the PV at the porta hepatitis of 11 dogs with IMHA and 21 clinically normal control dogs. Peak velocity did not differ significantly (P = 0.27) between groups. See Figure 3 for remainder of key.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
A receiver operator characteristic curve was used to quantify the predictive power of blood flow velocity in the CVC to identify patients with suspected IMHA. Area under the curve was 0.84 (Figure 5). The predictive ability of blood flow velocity in the CVC to identify patients with IMHA was assessed by use of a logistic regression model (Figure 6). Velocities ≥ 90.12 cm/s indicated a ≥ 50% probability of IMHA. At a velocity of 90.12 cm/s, sensitivity was 55% and specificity was 95%, as indicated on the receiver operator characteristic curve.
Receiver operating characteristic curve of the peak blood flow velocity in the CVC and its use as a tool to predict the presence of IMHA in dogs suspected of having the disease. The value for the area under the curve is 0.84.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
Results of a logistic regression model to determine the probability of a dog having IMHA on the basis of peak velocity in the CVC of dogs with suspected IMHA.
Citation: American Journal of Veterinary Research 79, 5; 10.2460/ajvr.79.5.518
Discussion
Analysis of the results reported here indicated that there were significant elevations in peak blood flow velocity in the CVC in the cranial portion of the abdomen of dogs affected with IMHA. This supported findings from a previous study5 in which severe and moderate anemia resulted in an increase in mean flow velocity over time in abdominal arteries (including the aorta, cranial mesenteric artery, celiac artery, and hilar splenic artery) and the main PV.5 However, blood flow velocity in the CVC was not evaluated in that study.5 To our knowledge, the study reported here represented the first quantitative evaluation of blood flow velocity in the CVC of dogs by use of pulsed-wave Doppler ultrasonography. Pulsed-wave Doppler ultrasonography was found to be an effective method for use in predicting the presence of IMHA by use of blood flow velocity in the CVC. Despite the challenge of obtaining diagnostic waveforms during respiratory motion in unsedated patients in the present study, the waveforms consistently correlated with the ECG tracings; there was a definitive antegrade S wave representing blood flow toward the heart, and smaller retrograde waves were visible intermittently. Antegrade D waves were less definitely identified than were S waves because of translational motion during respiration. Overall, these qualitative findings are similar to hepatic vein waveforms in clinically normal dogs9,10 and humans.11,12
Several reasons are possible for the fact that the peak velocity of the CVC was elevated in the clinically affected patients of the present study even though there was no significant difference in blood flow velocity in the PV between groups. First, flow through the cranial abdominal vena cava is subject to the balance of pressures between the right atrium and abdominal veins. Elevated hepatic venous S wave velocity has been reported in anesthetized dogs with induced hyperdynamic circulation and subsequent elevations in cardiac contractility, heart rate, cardiac output, and systemic arterial blood pressure.9 Volume status of the dogs with IMHA in the study reported here was considered equal to that of the clinically normal dogs and was normovolemic with no evidence to suggest otherwise, given the lack of tachycardia and the lack of difference in heart rates between groups. However, contractility of the myocardium in the IMHA group was likely increased owing to the presence of anemia, which created a need for a higher cardiac output. However, caution should be used in interpretation of the volume status because central venous pressures were not measured in these patients.
Altered vascular resistance attributable to anemia is another proposed cause of higher flow velocity. Severe anemia reportedly causes alterations in splanchnic blood distribution, with a decrease in vascular resistance to vital organs.5,14 A decrease in vascular resistance was further supported by the lack of significant alterations in venous diameter. The decrease in blood viscosity in dogs with IMHA may have played a large role in decreasing vascular resistance and, ultimately, increasing flow velocity. Experimentally inducing anemia in dogs can cause a decrease in viscosity that results in a large increase in cardiac output.15 In vitro studies6,16,17 of various blood viscosities and aggregating blood cells have found that the shape of the velocity profile within a blood vessel is highly correlated with viscosity and that aggregating blood cells result in local disturbances of blood flow, with higher velocities toward the wall where there are regions of increased shear stress and large aggregates of RBCs in areas of lower flow velocities and high RBC concentrations.
Finally, given the complexity of the disease process of IMHA, potential factors (eg, endothelial activation and active fibrinolysis) caused by the dysregulation in hemostasis may contribute to an increase in velocity.1 These factors warrant further study.
The reason that velocity in the PV was not significantly elevated is not known. This may have been attributable to a type 2 error, given that median velocity in the affected group was elevated (but not significantly different), compared with velocity in the clinically normal dogs. It is also possible that velocity in the PV was subjected to different local conditions with regard to viscosity and vascular resistance, compared with local conditions in the CVC. In dogs, the spleen has the greatest capacity to release blood into the circulation, which causes a local elevation in viscosity.18–20 The splenic vein contributes a large portion of blood to the PV.19 It is possible this local increase in viscosity may have counteracted other factors that would have contributed to velocity elevations in these patients.
Similar to results of previous studies,2,3,21 all the dogs with IMHA in the present study had a high plasma fibrinogen concentration. This is a marker of systemic inflammation (an acute-phase inflammatory protein), and the elevation in concentration was compatible with the fact that the affected dogs were in the acute phase of the disease. All the dogs with IMHA, except for 1, fulfilled the criteria of hypercoagulability. That dog met other inclusion criteria for the study and was not excluded. Interestingly, that dog had the highest blood flow velocity in the CVC of all dogs with IMHA, and it died soon after data acquisition. It has been reported22,23 that dogs with IMHA with a hypocoagulable or normocoagulable status have a worse outcome because of consumption of coagulation factors and platelets as a result of macroscopic thrombosis or disseminated intravascular coagulation. The presence of a hypercoagulable TEG tracing must be interpreted with caution in patients with IMHA because nonhemolytic anemias with an associated decrease in viscosity as well as hyperfibrinogenemia have also reportedly resulted in a hypercoagulable TEG tracing.24–26 Additionally, most of the dogs with IMHA had been receiving, at the discretion of the primary clinician, corticosteroids or other immunosuppressive treatments at the time of data acquisition. Corticosteroid treatment has been associated with a hypercoagulable state, increased MA on TEG, and decreased clot lysis.27,28 In the present study, when dogs with IMHA required blood products, care was taken to obtain TEG measurements before infusion of blood products and ultrasonographic measurements before or at the time of infusion.
The study reported here had a number of limitations, the most substantial of which was the small sample size of clinically affected dogs. Few eligible dogs were examined during the recruitment period; however, it is unlikely that the low number caused important errors, given the strong association for the difference in blood flow velocity in the CVC between groups. It would have been optimal to have obtained both ECG and central venous pressures on all patients to evaluate the influence of pressure alterations on flow velocity. Because of financial constraints as well as the unsedated status of patients, these procedures were not feasible. Radiography and complete physical examinations were performed to exclude right-sided heart failure, and abdominal ultrasonography was used to confirm a lack of hepatic venous distension and ascites.
Finally, there may have been potential errors in velocity measurements attributable to cursor placement in the vessel of interest and patient movement because of the lack of sedation and presence of respiratory motion. Efforts were made to ensure diagnostic images were acquired with careful placement of the cursor parallel to the vessel so that the angle of insonation was maintained between 30° and 60°, which reportedly is the optimal range of angle correction to minimize errors in velocity determination.8,29–32 This included careful placement of the cursor for measurements of diameter in the transverse plane. It is possible that slight oblique orientations of the probe can result in an overestimate of vascular diameter. It is not known whether multiple observers recording ultrasonographic measurements would alter the repeatability and reliability of the measurements.
Patient motion and respiratory motion caused slight spectral broadening sufficient to obscure the definition of a D wave, which inhibited evaluation of S wave-to-D wave ratios for the CVC. The S wave-to-D wave ratios have been described in clinically normal dogs and reportedly are lower in dogs with tricuspid regurgitation as a result of a decrease in systolic flow.9,33 It is unknown whether the S wave-to-D wave ratio in dogs with IMHA differs substantially from that of clinically normal dogs. Given the increase in S wave velocity, it is possible that the S wave-to-D wave ratio may have been increased in the dogs with IMHA in the present study, and this is an area for potential future study.
The study reported here was designed to investigate the influence of naturally occurring IMHA on blood flow velocity in main abdominal veins. Peak flow velocity toward the heart (determined by use of pulsed-wave Doppler ultrasonography) in the CVC in the cranial portion of the abdomen was significantly higher in dogs with IMHA, compared with the velocity for clinically normal dogs. Pulsed-wave Doppler ultrasonography is an effective modality for use in predicting anemia in patients suspected of IMHA. Further studies are warranted to evaluate the relationship between peak velocity in the CVC and response to treatment in hypercoagulable dogs as well as the relationship of the S wave-to-D wave ratio in the CVC on the pulsed-wave Doppler ultrasonographic waveform.
Acknowledgments
Supported by the Michigan Animal Health Fund. Funding sources did not have any involvement in the study design, data analysis and interpretation, or writing and publication of the manuscript.
The authors declare that there were no conflicts of interest.
ABBREVIATIONS
| CVC | Caudal vena cava |
| IMHA | Immune-mediated hemolytic anemia |
| MA | Maximum amplitude |
| PV | Portal vein |
| TEG | Thromboelastography |
Footnotes
Hemostasis analyzer 5000, Haemonetics, Niles, Ill.
GE Logiq S8, GE Medical Systems, Milwaukee, Wis.
R, version 1.0.44, R Core Team, Auckland, New Zealand.
R package pROC, version 1.8, R Core Team, Auckland, New Zealand.
Excel, version 12.3.6, Microsoft Corp, Redmond, Wash.
References
1. Kidd L, Mackman N. Prothrombotic mechanisms and anticoagulant therapy in dogs with immune-mediated hemolytic anemia. J Vet Emerg Crit Care (San Antonio) 2013;23:3–13.
2. Scott-Moncrieff JC, Treadwell NG, McCullough SM, et al. Hemostatic abnormalities in dogs with primary immune-mediated hemolytic anemia. J Am Anim Hosp Assoc 2001;37:220–227.
3. Fenty RK, Delaforcade AM, Shaw SE, et al. Identification of hypercoagulability in dogs with primary immune-mediated hemolytic anemia by means of thromboelastography. J Am Vet Med Assoc 2011;238:463–467.
4. Koma LM, Spotswood TC, Kirberger RM, et al. Influence of normovolemic anemia on Doppler characteristics of the abdominal aorta and splanchnic vessels in Beagles. Am J Vet Res 2005;66:187–195.
5. Taylor GA, Hudak ML. Color Doppler ultrasound of changes in small vessel diameter and cerebral blood flow during acute anemia in the newborn lamb. Invest Radiol 1994;29:188–194.
6. Grabowski EF. Thrombolysis, flow, and vessel wall interactions. J Vasc Interv Radiol 1995;6:25S–29S.
7. Lamb CR, Mahoney PN. Comparison of three methods for calculating portal blood flow velocity in dogs using duplex-Doppler ultrasonography. Vet Radiol Ultrasound 1994;35:190–194.
8. Szatmári VI, Sótonyi P, Vörös K. Normal duplex Doppler waveforms of major abdominal blood vessels in dogs: a review. Vet Radiol Ultrasound 2001;42:93–107.
9. Smithenson BT, Mattoon JS, Bonagura JD, et al. Pulsed-wave Doppler ultrasonographic evaluation of hepatic veins during variable hemodynamic states in healthy anesthetized dogs. Am J Vet Res 2004;65:734–740.
10. Nelson NC, Drost WT, Lerche P, et al. Noninvasive estimation of central venous pressure in anesthetized dogs by measurement of hepatic venous blood flow velocity and abdominal venous diameter. Vet Radiol Ultrasound 2010;51:313–323.
11. Abu-Yousef MM. Normal and respiratory variations of the hepatic and portal venous duplex Doppler waveforms with simultaneous electrocardiographic correlation. J Ultrasound Med 1992;11:263–268.
12. Coulden RA, Lomas DJ, Farman P, et al. Doppler ultrasound of the hepatic veins: normal appearances. Clin Radiol 1992;45:223–227.
13. Koenigshof AM, Scott MA, Brown AJ. Effects of delayed anticoagulation and use of evacuated tubes on non-activated thrombelastography in dogs. Vet Clin Pathol 2012;41:63–70.
14. Koma LM, Kirberger RM, Scholtz L, et al. Influence of normovolemic anemia on Doppler-derived blood velocity ratios of abdominal splanchnic vessels in clinically normal dogs. Vet Radiol Ultrasound 2005;46:427–433.
15. Fowler NO, Holmes JC. Blood viscosity and cardiac output in acute experimental anemia. J Appl Physiol 1975;39:453–456.
16. Sherwood JM, Kaliviotis E, Dusting J, et al. Hematocrit, viscosity and velocity distributions of aggregating and non-aggregating blood in a bifurcating microchannel. Biomech Model Mechanobiol 2014;13:259–273.
17. Yeom E, Kang YJ, Lee SJ. Changes in velocity profile according to blood viscosity in a microchannel. Biomicrofluidics 2014;8:034110.
18. Chien S, Dellenback RJ, Usami S, et al. Blood volume, hemodynamic, and metabolic changes in hemorrhagic shock in normal and splenectomized dogs. Am J Physiol 1973;225:866–879.
19. Carneiro JJ, Donald DE. Blood reservoir function of dog spleen, liver, and intestine. Am J Physiol 1977;232:H67–H72.
20. Pozniak MA, Kelcz F, Stratta RJ, et al. Extraneous factors affecting resistive index. Invest Radiol 1988;23:899–904.
21. Mischke R. Hemostatic disorders as a complication of autoimmune hemolytic anemia in dogs. Dtsch Tierarztl Wochenschr 1998;105:13–16.
22. Sinnott VB, Otto CM. Use of thromboelastography in dogs with immune-mediated hemolytic anemia: 39 cases (2000–2008). J Vet Emerg Crit Care (San Antonio) 2009;19:484–488.
23. Wiinberg B, Jensen AL, Johansson PI, et al. Thromboelastographic evaluation of hemostatic function in dogs with disseminated intravascular coagulation. J Vet Intern Med 2008;22:357–365.
24. Brooks AC, Guillaumin J, Cooper ES, et al. Effects of hematocrit and red blood cell-independent viscosity on canine thromboelastographic tracings. Transfusion 2014;54:727–734.
25. Smith SA, McMichael MA, Gilor S, et al. Correlation of hematocrit, platelet concentration, and plasma coagulation factors with results of thromboelastometry in canine whole blood samples. Am J Vet Res 2012;73:789–798.
26. Wagg CR, Boysen SR, Bedard C. Thrombelastography in dogs admitted to an intensive care unit. Vet Clin Pathol 2009;38:453–461.
27. Flint SK, Abrams-Ogg AC, Kruth SA, et al. Independent and combined effects of prednisone and acetylsalicylic acid on thromboelastography variables in healthy dogs. Am J Vet Res 2011;72:1325–1332.
28. Patrassi GM, Sartori MT, Livi U, et al. Impairment of fibrinolytic potential in long-term steroid treatment after heart transplantation. Transplantation 1997;64:1610–1614.
29. Rubens DJ, Bhatt S, Nedelka S, et al. Doppler artifacts and pitfalls. Radiol Clin North Am 2006;44:805–835.
30. Finn-Bodner ST, Hudson JA. Abdominal vascular sonography. Vet Clin North Am Small Anim Pract 1998;28:887–942.
31. Gill RW. Measurement of blood flow by ultrasound: accuracy and sources of error. Ultrasound Med Biol 1985;11:625–641.
32. Moriyasu F, Ban N, Nishida O, et al. Clinical application of an ultrasonic duplex system in the quantitative measurement of portal blood flow. J Clin Ultrasound 1986;14:579–588.
33. Kim J, Kim S, Eom K. Pulsed-wave Doppler ultrasonographic evaluation of hepatic vein in dogs with tricuspid regurgitation. J Vet Sci 2017;18:73–79.
