Abdominal Chronic Expanding Hematoma Causing Iron-Deficiency Anemia in a Dog

Case Report

A 2 yr old spayed female mixed-breed Irish wolfhound weighing 27.2 kg presented to the Veterinary Health Center at Kansas State University for evaluation of abdominal distension secondary to hemoabdomen. The dog was initially examined by the primary care veterinarian for routine examination and ovariohysterectomy 10 wk earlier. Because the dog had been adopted from a rescue organization a few days before that examination and ovariohysterectomy, the dog’s history was unknown. Physical examination was reportedly unremarkable; however, at the time of ovariohysterectomy, approximately 2 L of yellow-tinged peritoneal fluid was noted and the uterus was adhered to the colon and jejunum. The adhesions were broken down, the fluid was removed, and the ovariohysterectomy was completed. No analysis of the peritoneal fluid was conducted, and additional diagnostics were declined at that time.

Ten days later, the patient presented to the referring veterinarian for reported polyuria, polydypsia, and anorexia. A complete blood cell count (CBC) and serum biochemical profile were reported to be within normal limits, and the patient was started on a short course of trimethoprim/sulfamethoxazole (18 mg/kg per os [PO] q 12 hr for 1 wk). One month prior to presentation to the Veterinary Health Center, moderate abdominal distension was noted by the owner, which continued to worsen until the patient was examined by the primary care veterinarian 11 days prior to presentation to the Veterinary Health Center. The dog was reported to be lethargic, tachycardic, and had a moderate abdominal distention. A CBC and serum biochemical profile performed at that time reportedly showed that the dog was anemic and panhypoproteinemic. Abdominal radiographs revealed a pendulous abdominal contour and poor serosal detail. Abdominocentesis yielded approximately 3.5–4 L of red-tinged peritoneal fluid, with few red blood cells and occasional neutrophils on microscopic examination; however, no quantitative analysis of the fluid was conducted. The patient was discharged with prednisone (0.4 mg/kg PO q 12 hr) and ciprofloxacin (18 mg/kg PO q 12 hr) for 1 wk. Four days later, the patient was examined for continued abdominal distention and tenesmus and an enema was given. The dog was subsequently referred to the Veterinary Health Center for further evaluation.

At the time of presentation to the Veterinary Health Center, the dog was thin (body condition score, 1 out of 5) and had moderate generalized muscle atrophy. Mucous membranes were pale. The abdomen was distended and turgid with a notable fluid wave. Thoracic auscultation showed tachycardia. Peripheral pulses were difficult to palpate, and the patient’s distal limbs were cool to the touch. Rectal examination was unremarkable.

A CBC revealed a marked microcytic, hypochromic, regenerative anemia (packed cell volume was 0.12; reference range, 0.37–0.55), erythrocyte concentration was 2.14 × 10¹²/L (reference range, 5.5–8.5 × 10¹²/L), mean corpuscular volume was 55 fL (reference range, 60–77 fL); mean corpuscular hemoglobin concentration was 270 g/L (reference range 320–360 g/L ); reticulocyte percentage was 0.132), and thrombocytosis (platelet count was 1294 × 10⁹/L; reference range, 164–510 × 10⁹/L). The serum biochemical profile revealed mild hypoalbuminemia (31 g/L; reference range, 34–42 g/L) and hypocholesterolemia (2.87 mmol/L; reference range, 3.44–10.20 mmol/L). The prothrombin time was 8.1 sec (reference range, 7.1–9.1 sec) and the activated partial thromboplastin time was 11.6 sec (reference range, 8.2–12.7 sec). A urinalysis was unremarkable, with a urine specific gravity of 1.022.

An abdominocentesis was performed shortly after presentation and 2 L of hemorrhagic fluid was removed. The fluid was cloudy and blood tinged, with a packed cell volume of 0.05, a total nucleated cell count of 3.3 × 10⁹/L, and a protein level (by refractometry) of 50 g/L. Microscopic evaluation of the fluid was consistent with acute hemorrhage with a large number of erythrocytes, a moderate number of leukocytes, and few platelets. No organisms, inflammatory cells, or neoplastic cells were observed.

On thoracic radiographs, the heart, lungs, and vessels were unremarkable, but in the limited view of the abdomen there was loss of serosal detail and dorsal and caudal displacement of the stomach. An abdominal ultrasound revealed an extremely large cavitated structure, filling the majority of the abdominal cavity and displacing the abdominal organs cranially, caudally, and dorsally. It became clear at that point that all abdominocentesis attempts (by the referring veterinarian and the authors) were more likely yielding fluid from the large structure and not the peritoneal cavity. The mass was filled with echogenic fluid, and multiple linear echogenic septations were present within the mass (Figure 1). A small amount of anechoic peritoneal fluid was present, and the liver was subjectively decreased in size, being only visible from a right intercostal window. The spleen, kidneys, urinary bladder, and gastrointestinal tract were normal in appearance, despite being displaced.

View Figure

• Download as Powerpoint
• Download Figure

Due to the suspicion of iron deficiency anemia (based on red blood cell indices) but without apparent external blood loss, an iron profile was submitted on serum and the iron concentration was measured in the fluid removed from the cavitated structure. Serum iron concentration was decreased at 7.34 µmol/L (reference range, 15.75–42.60 µmol/L), total iron binding capacity was normal at 80.55 µmol/L (reference range, 44.03–80.55 µmol/L), and the serum ferritin concentration was normal (220.2 pmol/L; reference range, 179.8–1797.6 pmol/L). The concentration of iron in the fluid from the cavitated structure was 178.9 µmol/L, which was >24 times greater than the serum iron concentration. That finding was consistent with iron-deficiency anemia (IDA) from chronic abdominal hemorrhage and sequestration of iron in the cavitated structure.

On day 1450 mL of whole blood (16.5 mL/kg) was administered IV over 4 hr, and iron dextran^a (20 mg/kg) was administered IM once. On day 2, an abdominal exploratory surgery was performed (Figure 2). The dog was sedated with midazolam^b (0.2 mg/kg subcutaneously [SC]) and hydromorphone^c (0.1 mg/kg SC). Anesthesia was induced with propofol^d (6 mg/kg IV to effect) and maintained with isoflurane^e. Mechanical ventilation was used during the entire procedure, with an intermittent positive-pressure ventilation of 12 cm water. Perioperative ampicillin Na/sulbactam Na^f (25 mg/kg IV) was administered q 90 min, and perioperative analgesia was established with ketamine^g (600 µg/kg/hr).

View Figure

• Download as Powerpoint
• Download Figure

A standard ventral midline incision was made from the xyphoid to pubis, and the abdomen was opened. Upon entry to the peritoneal cavity, a large fluid-filled structure was encountered, which seemed to fill the entire abdominal cavity (Figure 2A). An incision was made into the wall of the structure, and suction was used to drain 6.5 L of red wine-colored fluid from this structure (Figure 2B). Windows were made through that structure’s wall to visualize the rest of the abdominal cavity and slowly break down the numerous adhesions present between the cavitated structure and the abdominal organs (Figure 2C). Adhesions were broken down, and the structure was divided into cranial and caudal portions. The cranial portion was separated from the abdominal organs and around the liver to the level of the hepatic hilum. The structure was incised at that point, and a small portion of this cranial part of the structure was left in the abdomen so that other structures, such as the biliary tree, the caudal vena cava, and the esophagus, were not disrupted. The caudal portion of the structure was then separated from the abdominal organs to the level of the bladder, and a stalk of tissue was found to be traveling from the cavitated structure to the right side of the body wall (Figure 2D). That tissue was incised, and the caudal portion of the structure was freed from the abdomen. Portions of the structure were submitted for culture (aerobic and anaerobic) and histologic examination. In addition, a liver biopsy was obtained for histological examination. The liver biopsy showed irregular patchy areas of hepatocellular vacuolization.

Even after removal of the structure, the right kidney, ureters, bile duct, and a majority of the intestinal tract could not be clearly identified because of significant adhesion formation (Figure 2E). Abdominal closure was routine, and recovery from anesthesia was uneventful. The packed cell volume obtained immediately prior to surgery was 0.33 (total solids, 48 g/L), but both had dropped (PCV, 0.20; total solids, 24 g/L) when reassessed shortly after recovery from anesthesia. Whole blood (430 mL IV over 4 hr) was administered. Postoperatively, analgesia was managed with fentanyl^h (3 µg/kg/hr). Additional therapies included heparinⁱ (95 U/kg SC q 8 hr), ampicillin/sulbactam (25 mg/kg IV q 8 hr), tramadol^j (5.9 mg/kg PO q 8 hr), and ferrous sulfate^k (7.7 mg/kg PO q 12 hr).

On day 3 of admittance (i.e., day 1 postsurgically), moderate abdominal distension was noted and abdominocentesis was performed. In total, 380 mL of a blood-tinged, cloudy fluid was removed. Analysis of that fluid showed a protein level of 26 g/L (by refractometry), a packed cell volume <0.03, and a total nucleated cell count of 3.8 × 10⁹/L that consisted primarily of nondegenerate neutrophils (>95% of cells), activated macrophages, and occasional small lymphocytes and reactive mesenchymal cells. The dog was otherwise stable and eating; therefore, no change in the treatment protocol was initiated pending results of the histologic evaluation of submitted tissue. The dog was discharged on day 5 with amoxicillin trihydrate/clavulanate potassium^l (17 mg/kg PO q 12 hr for 3 wk), tramadol (5.9 mg/kg PO q 8 hr for 1 wk), and ferrous sulfate (7.7 mg/kg PO q 12 hr for 3 wk). The packed cell volume at the time of discharge was 0.35 (total solids, 30 g/L).

The dog was euthanized by the primary care veterinarian 12 days after discharge from the Veterinary Health Center for progressive lethargy and abdominal distension. The body was submitted to the Veterinary Diagnostic Laboratory at Kansas State University for postmortem examination 48 hr after being euthanized.

Microscopic Examination

All tissue samples were fixed in 10% neutral buffered formalin. The samples were routinely processed, paraffin-embedded, sectioned at 5 µm, and stained with hematoxylin and eosin. Histologically, the cavitated structure removed at surgery was not lined by an epithelium, so it could not be a considered a cyst but rather as a pseudocyst. The lumen contained anastomosing trabeculae composed of eosinophilic lamellated amorphous material that multifocally surrounded eosinophilic fibrillar material (necrosis and organizing fibrin). Some sections contained large numbers of extravasated erythrocytes (hemorrhage) in the lumen (Figure 3). The pseudocyst was rimmed by thick collagenous tissue (capsule), loosely arranged reactive fibroblasts and small caliber blood vessels lined by reactive endothelial cells that were frequently arranged perpendicular to the surrounding fibrous connective tissue (granulation tissue) and small numbers of lymphocytes, plasma cells, and hemosiderin-laden macrophages (Figure 3). The outer surface of the structure was covered by thick mats of fibrin, hemorrhage, abundant cellular debris, and small numbers of neutrophils. Neither organisms nor neoplastic cells were observed. The gross and histologic findings were consistent with chronic expanding hematoma (CEH).^1–12

View Figure

• Download as Powerpoint
• Download Figure

The organs and tissues that were obtained at necropsy (rectus abdominis, diaphragm, liver, stomach, small and large intestines, pancreas, kidneys, urinary bladder, lymph nodes, spleen, and adrenal glands) were normal microscopically, except for a 1 mm thick layer of granulation tissue lacking epithelium with multifocal infiltrates of low numbers of macrophages, lymphocytes, and plasma cells adhered to the serosal surface of all of those organs.

Discussion

The unusual aspects of the case described in this report were the CEH of the peritoneal cavity and IDA secondary to iron sequestration in the CEH. The histological features of the pseudocyst in the dog of this report exhibited the findings of CEH in humans and the few reports of CEH in other animals.^1–12

In humans, CEH appear as a firm, painless, circumscribed, slowly expanding mass, growing sometimes over years and often simulating neoplasms.^1,2,6 Histopathologically, a CEH consists of a thick fibrocollagenous outer layer and an inner granulation layer surrounding the hematoma cavity.¹ Such hematomas have been described in variety of locations (e.g., brain, muscle, subcutaneous, pelvic, and thoracic and abdominal cavities) and are presumed to occur following surgery, trauma, or injury; however, an inciting cause has not been identified in every case.^1–7

Unlike classic hematomas that usually stabilize quickly by equalizing IV and extravascular pressure rapidly, CEH continually expand. The mechanism whereby exudation or bleeding continues within a fibrous capsule remains unknown, but is believed to have an inflammatory cause.^1,13 Breakdown products derived from erythrocytes, hemoglobin, leukocytes, and other solid blood elements induce mild, continuous inflammation, causing an increased permeability of the vascular wall and bleeding from capillaries in the granulation tissue, resulting in expansion of the hematoma.^1,13 Activation of the vascular endothelial growth factor pathway may lead to neovascularization in the granulation tissue layer of the hematoma capsule that serves as a continual source of bleeding and perifocal edema.⁷ Multiple septations are a common finding of CEH, which may suggest chronic expansion of different stages caused by recurrent bleeding into the existing hematoma.

In veterinary medicine, cases of CEH have been reported in dogs, a cat, a horse, and a bull, with various locations (e.g., subcutaneous tissue, perirenal area, ovary, penis).^8–12 In the five dogs reported in the literature, all were <1 yr of age and potential trauma (vaccination, scruffing) was reported as a possible inciting cause in all.⁸ Likewise, the cat (intramuscular injection suspected of puncturing the kidney), the horse (failed attempt to jump a fence), and the bull (breeding) had suspected traumatic events.^10–12 The dog in this report was adopted from a shelter, and the history prior to the current owner was unknown. A coagulopathy was an unlikely cause of the hematoma in this case given that the coagulation profile was within reference ranges, making unknown trauma the most likely etiology. Although thrombopathia and von Willebrand disease were not evaluated as possible causes of the CEH, there was no excessive capillary hemorrhage associated with paracentesis, venipuncture, or surgical incision.

The preferred treatment of CEH in humans is surgical excision. The goals of surgery are complete resection of the capsule and elimination of the dead space (to prevent reoccurrence).^1–7 In veterinary medicine, en bloc resection of the mass is also the recommended treatment.^8–10,12 Aspiration of the fluid followed by the use of a pressure bandage has given poor results.^8,9 Repeated aspiration is also likely to cause more hemorrhage and blood loss because the inner side of the wall is vascularized.

Surgical excision gives good results in most cases. No recurrence was noticed in five dogs with cervical hematomas and a bull with penile hematoma.^8,12 Recurrence can occur if the entire capsule has not been removed, which is suspected to have happened in the dog in this report.³ The numerous adhesions to the body wall and all the abdominal organs made complete resection difficult and it was accepted that a small portion of the structure was left behind. The placement of an abdominal drain did not appear indicated at time of surgery, since it was a sterile process and the nature of the mass was unknown.

Given the degree of dissection that was required to remove the mass, the study authors’ concern was that the degree of vascular and microvascular trauma could have predisposed the dog to disseminated IV coagulopathy, prompting thromboprophylaxis with heparin. The hemorrhage was occurring within the CEH that was removed. Therefore, the authors’ concern for ongoing hemorrhage was low.

To the authors’ knowledge, IDA has not previously been reported secondary to sequestration of iron within the body. Although IDA can result from either inadequate dietary intake, inadequate absorption of iron, or chronic blood loss, the most common cause is chronic external blood loss associated with gastrointestinal bleeding, usually secondary to neoplasia, gastric ulcers, or severe endoparasitism.¹⁴ Other causes of chronic external blood loss include heavy flea infestation, urogenital bleeding, and iatrogenic bloodletting.¹⁴ Internal blood loss (such as hemoabdomen) does not usually lead to IDA because iron is typically reabsorbed by the body. For the dog in this report, iron was presumptively trapped within the large cavitary structure and isolated from an absorptive surface, leading to iron depletion. Indeed, the iron concentration in the fluid collected from the cavitated structure was >24 times greater than the serum iron concentration (999.2 µg/dL versus 41 µg/dL). Although blood loss alone into the CEH could have resulted in a low hematocrit, it should not have resulted in either microcytosis or hypochromia, both of which are hallmarks of IDA.

Conclusion

This report details a rare condition, CEH, in the peritoneal cavity of a dog with secondary IDA from sequestration of iron within the CEH. Due to the difficulty in complete resection of the CEH in this dog and persistent abdominal hemorrhage from the residual tissue, the outcome of this case was unfavorable. However, prognosis of future cases may depend on the size of the CEH, its location, and the ability (or not) to fully remove the structure surgically. CEH should be considered a primary differential diagnosis when an expanding mass (abdominal or elsewhere) and IDA occur in concert without evident external blood loss.