• Open Access

Right Ventricular and Saphenous Vein Thrombi Associated with Sepsis in a Quarter Horse Foal


Corresponding author: Heidi Banse, DVM, DACVIM, 1 BVMTH, Veterinary Clinical Sciences, Oklahoma State University, Stillwater, OK 74078-2041; e-mail: heidi.banse@okstate.edu.


activated partial thromboplastin time


disseminated intravascular coagulation


immunoglobulin G


prothrombin time


pulmonary thromboembolism

A 41-kg, 12-hour-old Quarter Horse filly presented to the Boren Veterinary Medical Teaching Hospital for a 12-hour history of weakness and recumbency. The mare had a history of premature lactation and presumptive placentitis, which was treated with trimethoprim-sulfamethoxazole by the referring veterinarian. The filly was born at 315 days gestation; parturition was unattended. The filly was found recumbent and unable to stand. Before referral, the attending veterinarian administered 1 L of colostrum via nasogastric tube. Upon presentation to the veterinary teaching hospital, the filly was recumbent and lethargic with intermittent bouts of hyperresponsiveness. Rectal temperature was 94°F, heart rate was 132 beats per minute, and respiratory rate was 24 breaths per minute. A grade 3/5 systolic murmur was auscultated over the left heart base. Mucous membranes were hyperemic with a capillary refill time of approximately 2 seconds. Abrasions were present along the gingiva and around the eyes and muzzle. Petechiae were observed in both pinnae. The foal's distal limbs were cool and edematous, peripheral pulses were weak, and the coronets of all 4 feet were circumferentially dark red.

Presumptive diagnoses of neonatal encephalopathy and sepsis were made. An IV jugular catheter was placed. Blood was collected for bacterial culture and sensitivity, CBC, fibrinogen concentration, serum biochemistry, and quantification of serum immunoglobulin G (IgG) concentration. The CBC disclosed immature (band) neutrophilia (558 bands/μL; reference range, <200 bands/μL) and toxic polymorphonuclear leukocytes. A biochemistry profile identified hypoproteinemia (3.8 g/dL; reference range, 5.6–8.0 g/dL) characterized by hypoglobulinemia (1.1 g/dL; reference range, 2.6–5.6 g/dL), azotemia (creatinine, 4 mg/dL; reference range, 1.2–2.0 mg/dL), hypoglycemia (37 mg/dL; reference range, 70–120 mg/dL), hypernatremia (156 mEq/L; reference range, 130–146 mEq/L), hyperchloremia (112 mEq/L; reference range, 95–110 mEq/L), and increased creatine kinase activity (40016 IU/L; reference range, 20–500 IU/L). Quantification of IgG1 concentration was consistent with failure of passive transfer (IgG < 400 mg/dL). Neonatal sepsis score was calculated to be 18.[1] Arterial blood gas sampling was not attempted because of limb edema and weak peripheral pulses. Blood pressure measured using an oscillometric blood pressure monitor2 applied to the middle coccygeal artery yielded a mean arterial pressure of 70 mmHg.

Initial medical treatment consisted of a plasma3 transfusion (25 mL/kg), IV fluids (1 L bolus of 0.45% sodium chloride followed by 0.45% sodium chloride with 10% dextrose at 6 mg/kg/min), broad spectrum antimicrobials (ceftiofur,4 10 mg/kg IV q6h), dimethyl sulfoxide5 (1 g/kg/d diluted to a 10% solution IV), thiamine (10 mg/kg IV q24h), and magnesium sulfate (25 mg/kg/h). After plasma3 transfusion, quantification of serum IgG1 concentration indicated persistent hypogammaglobulinemia (IgG, 400–800 mg/dL); an additional plasma3 transfusion (25 mL/kg) was administered resulting in adequate IgG concentrations (>800 mg/dL). Initial urine specific gravity (after approximately 3 L of IV fluids) was 1.025. The foal improved with treatment and was able to stand with assistance within 24 hours of admission. Pan feeding of the mare's milk was implemented at a rate of 5% body weight per day; the volume was gradually increased over 4 days until the filly was nursing free choice.

Serum creatinine concentration decreased after 12 hours of fluid therapy (3.6 mg/dL), but urine output was decreased and clinical evidence of peripheral and pulmonary edema developed. Urine specific gravity then became isosthenuric (1.010–1.017). Constant rate infusions of furosemide6 (0.12 mg/kg bolus followed by 0.12 mg/kg/h) and dopamine (2.5 μg/kg/min) were initiated to improve renal blood flow and urine output. Pulmonary edema resolved and peripheral edema improved. Edema of the right hind limb (from stifle to coronet) persisted. A biochemistry profile on Day 2 disclosed marked hyperglycemia (342 mg/dL) with normal insulin concentration (17.1 μU/mL; mean ± standard error of the mean for healthy foals 24–48 hours of age, 17.9 ± 3.4 μU/mL)2 and hyperlipemia (triglycerides, 1137 mg/dL; reference range, 25–120 mg/dL). Azotemia (serum creatinine concentration, 2.8 mg/dL) continued to improve. Serial measurements of blood glucose concentration indicated gradually increasing concentrations (from 125 mg/dL 24-hour postadmission to 483 mg/dL 36-hour postadmission) that persisted above 250 mg/dL for 12 hours despite discontinuation of IV dextrose supplementation. A constant rate infusion of insulin7 (0.07 IU/kg/h) was initiated, resulting in normalization of blood glucose concentration.

Ultrasound examination of the right hind limb was performed on Day 2 because of persistent, progressive limb edema. A thrombus was visible within the saphenous vein, at the level of midgaskin. Color flow Doppler confirmed absence of blood flow within the saphenous vein distal to the thrombus. A coagulation profile indicated prolonged prothrombin (PT) and activated partial thromboplastin times (aPTT) and hyperfibrinogenemia (Table 1). Clinical findings of venous thrombosis coupled with prolonged coagulation times supported a diagnosis of disseminated intravascular coagulation (DIC). The absence of clinical evidence of hemorrhage and the presence of a saphenous thrombus were consistent with a procoagulant state. Therefore, anticoagulant treatment with warfarin (0.018 mg/kg PO q24h for 5 days) and aspirin (20 mg/kg PO q48h) was initiated. Serial biochemistry profiles and urine specific gravity measurements identified resolution of azotemia and normalization of urine specific gravity (<1.008) by Day 4; furosemide6 and dopamine were discontinued. Forty-eight hours after implementation of anticoagulant treatment, ultrasound examination of the right hind limb identified a patent saphenous vein. Serial blood glucose concentrations were within normal limits and hypertriglyceridemia resolved by Day 5, and the constant rate infusion of insuling was discontinued. Blood culture results yielded mixed growth of Acinetobacter lwoffii and Streptococcus spp; antimicrobial treatment was changed to amikacin8 (25 mg/kg IV q24h) and penicillin9 (25,000 IU/kg IV q6h) based on sensitivity results.

Table 1. Coagulation parameters. Reference ranges are for adult horses.
Parameter (Reference Range)Day 2Day 6Day 8Day 11Day 17Day 21Day 28Day 49
  1. aPTT, activated partial thromboplastin time; PT, prothrombin; NA, not evaluated.

  2. a

    Control foal had an aPTT of 58.9 seconds and a PT of 9.8 seconds.

PT (9–12 seconds)13.812.813.512.811.311.91211.6a
aPTT (28–44 seconds)146.589.680.480.17169.557.263.5a
Fibrinogen (100–400 mg/dL)484192192238318140162160
d-dimer (<250 ng/mL)NA>2,0001,000–2,000>2,000500–1,000250–500<250<250
Platelets (100,000–400,000/μL)119,00075,000132,000290,000NA385,000NANA

Serial coagulation profiles indicated improved coagulation times and d-dimer concentrations, but aPTT remained prolonged (Table 1). A focal swelling developed at the site of catheter entry on Day 7; ultrasound examination of the jugular vein disclosed mild phlebitis, but no evidence of thrombus formation. On Day 8, the filly developed acute lameness (grade 4/5) in the right hind limb with a wound over the dorsolateral aspect of the metatarsophalangeal joint and associated joint effusion after being stepped on by the mare. Radiographs disclosed soft tissue swelling around the metatarsophalangeal joint. Lameness resolved after a single dose of firocoxib10 (0.2 mg/kg IV). Over the next several days, multiple similar wounds (well-circumscribed areas of necrotic skin 2–7 cm in diameter) were identified along all 4 limbs (Fig 1). Although some skin lesions were found over potential pressure points, multiple lesions were found in locations atypical for trauma secondary to recumbency (eg, dorsomedial metacarpus, medial stifle, groin). These lesions were presumed to be associated with microvascular thrombosis and dermal necrosis secondary to coagulopathy, although warfarin-induced skin necrosis could not be ruled out.[3] Skin lesions were treated with topical antimicrobials (silver sulfadiazine) and bandaging.

Figure 1.

(A) Photograph taken Day 10 of hospitalization depicting a well-circumscribed skin lesion on the medial aspect of the right metatarsus following sloughing of necrotic skin. 46 × 104 mm (150 × 150 DPI). (B) Photograph taken Day 9 of hospitalization depicting a well-circumscribed area of necrotic skin on the dorsal aspect of the left front fetlock. 136 × 181 mm (300 × 300 DPI).

On Day 15, the foal had an episode of marked tachypnea (100–120 breaths per minute) with nostril flare, although it remained bright and active. Attempts to obtain an arterial blood sample for blood gas analysis were unsuccessful. A CBC obtained the previous day had disclosed a normal white cell count and mild thrombocytosis (407,000 platelets/μL; reference range, 100,000–400,000 platelets/μL). The tachypneic episode resolved within an hour. Thoracic ultrasonography, radiography, and echocardiography were performed. Thoracic ultrasound examination identified mild comet-tail artifacts in the cranioventral and middle lung fields, bilaterally. No abnormalities were identified on thoracic radiographs. A 1 × 2 cm immobile thrombus was identified in the right ventricle on echocardiography (Fig 2). Anticoagulant treatment (warfarin, 0.018 mg/kg PO q24h, increasing dose by 20% q24h and heparin 150 IU/kg SC q12h for 24 h, then 125 IU/kg q12h for 4 days)[4] was initiated; aspirin was continued. Antimicrobial treatment was withheld for 24 hours. A blood culture was obtained to evaluate if the thrombus was septic; culture yielded no growth. On Day 19, antimicrobial treatment was changed to doxycycline based on blood culture results at admission; the IV catheter was removed. The foal was discharged on Day 24 with no change in size of the intracardiac thrombus. Aspirin and warfarin treatments were continued.

Figure 2.

Echocardiogram demonstrating an immobile right ventricular thrombus, identified Day 17 of hospitalization. Image was obtained using a 3.5 Mhz phased array probe in the 4th intercostal space, aimed slightly caudal from a standard right ventricular outflow tract view. Depth of field of view is 11 cm; the marks on the scale indicate 2 cm.

The foal was apparently healthy on evaluation on Day 49, but with persistence of a grade 2/5 systolic murmur over the left heart base. No abnormalities were identified on echocardiography; the intracardiac thrombus had resolved (Fig 3). A coagulation profile identified persistence of the prolonged aPTT (Table 1) compared with laboratory adult reference ranges, but an age-matched control foal had a similar aPTT. Treatments were discontinued.

Figure 3.

Echocardiogram demonstrating resolved right ventricular thrombus, Day 49. Image was obtained using a 3.5 MHz phased array probe in the 4th intercostal space, aimed slightly caudal from a standard right ventricular outflow tract view. Depth of field of view is 14 cm; the marks on the scale indicate 1 cm.

To the authors’ knowledge, this is the 1st report of an intracardiac thrombus in a foal. Although DIC occurs commonly in septic foals, reports of thrombi are rare.[5-7] Although the exact cause of thrombus formation in this case remains unknown, development most likely was secondary to an acquired coagulopathy associated with sepsis. Thrombus formation may occur after an inherited or acquired coagulopathy, mechanical irritation of the vascular endothelium, or alterations in blood flow.[8] Although coagulopathy is relatively common in septic human neonates,[9] overt thrombus formation is rare.[10] Compared with adults, human neonates exist in a procoagulant and antifibrinolytic state.[11] Similarly, neonatal foals have decreased antithrombin III and plasminogen and increased plasminogen activator inhibitor activities compared with adult horses.[12]

Sepsis may further exacerbate the prothrombotic state of the neonate. Sepsis is associated with increased platelet activation and decreased activity of anticoagulant and fibrinolytic systems. Approximately 50–70% of septic adult human patients have coagulation abnormalities and 35% are in DIC.[13] Although overt sepsis rarely occurs in the adult horse, coagulopathies occur during severe gastrointestinal disease.[14, 15] In contrast, coagulopathies are common in critically ill foals, with clinically detectable bleeding occurring in 47% of foals with sepsis or septic shock.[16] Coagulopathy secondary to systemic inflammatory response syndrome or sepsis is associated with high mortality in human neonates,[17] although this is not a consistent finding in foals.[18, 19]

In human neonates, small vessel size and frequent use of central venous lines may contribute to intracardiac thrombus formation.[8] In this foal, the catheter was within the jugular vein, rather than the right atrium. Although it is possible that the catheter tip created a thrombogenic environment, there was no evidence of thrombus formation at the tip of the catheter. It seems unlikely that the jugular catheter would have resulted in formation of an intracardiac thrombus without signs of marked inflammation or thrombus formation at the catheter site.

People with intracardiac thrombi are at increased risk for pulmonary thromboembolism (PTE)[20] and may present in respiratory distress. PTE was not confirmed in this case, but the bout of marked tachypnea led to investigation of the cardiopulmonary system and discovery of the intraventricular thrombus. In adult horses, the most common clinical sign associated with PTE is tachypnea.[21] Radiographs may support a diagnosis of PTE, but the diagnostic modality of choice is computed tomography pulmonary angiography.[22] Although additional diagnostic tests were declined, the clinical findings of a single episode of tachypnea coupled with normal thoracic radiographs, rectal temperature, and CBC and mild ultrasonographic changes support the possibility of PTE in this filly.

In people, thrombi (including intracardiac thrombi) often are removed surgically or dissolved by fibrinolytic treatment. In this foal, surgery was not an option because of expense and risk of the procedure. Fibrinolytic treatment was not pursued because of concerns about causing detachment of the thrombus from the right ventricle and migration into the pulmonary artery. Furthermore, intravascular fibrinolytic treatment previously has been reported to exacerbate coagulopathy in a foal.[6] Previous reports in human neonates have shown a low risk of embolization and frequent resolution with anticoagulant treatment alone.[23] Low molecular weight heparin is the most common anticoagulant used for thrombus resolution in human infants.[23] In this case, however, the aPTT already was substantially prolonged, consequently, anticoagulants that predominantly affect platelet aggregation (aspirin) and the extrinsic coagulation pathway (warfarin) were chosen. In this case, there was no evidence of sepsis at the time of thrombus detection and the foal's blood culture was negative. Because of the low sensitivity (approximately 60%)[24] of a single blood culture in septic foals, serial cultures often are recommended. However, the risk of withholding antimicrobial treatment longer than 24 hours was considered greater than the benefit of serial blood culture. Consequently, antimicrobials were reinstituted and conservative treatment with anticoagulants was chosen to treat the thrombus.

Although anticoagulant treatment frequently is used to treat thrombi in people, reports of spontaneous resolution exist. In this case, the saphenous vein thrombus resolved within 48 hours after initiating anticoagulant treatment. Coagulation times improved substantially over this time (PT by 7% and aPTT by 39%). These changes are likely a reflection of improving DIC rather than treatment. Aspirin has been demonstrated to prolong bleeding time for up to 48 hours without a clinically relevant effect on aPTT or PT in the adult horse.[25] Advanced platelet aggregation assays would have been useful in further evaluation of aspirin treatment, but were unavailable. Warfarin appeared to have no effect on PT in this case, and may not have hastened thrombus dissolution.

Because of the absence of age-specific reference ranges at our laboratory, interpretation of coagulation times is difficult. The persistently prolonged aPTT observed during hospitalization may have been a reflection of absence of age-specific reference ranges. The efficacy of anticoagulant treatment for resolution of thrombi in this foal remains unclear, but it seems likely that treatment of sepsis was the most important part of promoting a stable coagulation state and allowing passive dissolution of the thrombus.

This case report demonstrates that multiple thrombi, including an intracardiac thrombus, may occur in association with sepsis and DIC in a foal. In blood culture-negative human neonates, intracardiac thrombi had a mean time of resolution of 24 ± 24 days, with all cases resolving within 77 days.[23] The intracardiac thrombus in this foal resolved between 7 and 32 days from the date of detection. In conclusion, medical treatment including antimicrobials and supportive care may yield a successful outcome in cases of multiple thrombi associated with sepsis in foals. Although the usefulness of anticoagulants in the resolution of thrombi in foals remains undetermined, this case demonstrates that treatment with fibrinolytics may not be necessary for resolution of intracardiac thrombi.


This study was not supported by a grant or otherwise.


  1. 1

    SNAP Foal IgG Test, Idexx Laboratories, Westbrook, ME

  2. 2

    Veterinary Monitor 9403, Midmark Corporation, Versailles, OH

  3. 3

    ImmunoGlo, Mg Biologics, Ames, IA

  4. 4

    Naxcel, Pfizer Animal Health, New York, NY

  5. 5

    Domoso, Fort Dodge Animal Health, Fort Dodge, IA

  6. 6

    FuroJect, Butler Schein Animal Health, Dublin, OH

  7. 7

    Novolin R, Novo-Nordisk A/S, Bagsvaerd, Denmark

  8. 8

    Amiglyde-V, Fort Dodge Animal Health

  9. 9

    Pfizerpen, Pfizer Animal Health

  10. 10

    Equioxx, Merial Limited, Duluth, GA