Thirteen equids were initially examined from October 2001 to November 2008 for signs of colic (n = 7), weight loss (6), anorexia (3), or diarrhea (2). Colic signs were classified as either mild (n = 3) or severe (4). Duration of clinical signs ranged from 5 hours (colic) to 6 weeks (recurrent diarrhea). One horse, a 10-year-old Quarter Horse stallion, had a 3-week history of weight loss despite a normal appetite, generalized muscle wasting, and elevated digital arterial pulse strength in the forelimbs. Equids were admitted to hospitals at the University of Georgia (n = 6), Oklahoma State University (3), Marion duPont Scott Equine Medical Center (2), Auburn University (1), and the University of Missouri (1). There were 8 geldings, 4 mares, and 1 stallion. Age ranged from 1 to 18 years (median, 6 years). Equids represented included Quarter Horse (n = 4), American Paint Horse (2), donkey (2), Appaloosa (1), Haflinger (1), Tennessee Walking Horse (1), Thoroughbred crossbred horse (1), and pony (1). Ten equids were evaluated in fall (September to November), 2 were evaluated during winter (December to January), and 1 was evaluated in spring (March). In 7 equids, a history of access to persimmon was reported at the time of admission; all of these equids were evaluated in fall (September to October).
Physical examination abnormalities at the time of admission included tachycardia (> 48 beats/min) in 5 equids and tachypnea (> 20 breaths/min) in 3 equids. Mild increases in rectal temperature (ie, fever) were noted at admission in 3 equids (38.4° to 38.7°C [101.2° to 101.7°F]; median, 38.7°C). Three equids had > 2 L of gastric reflux upon passage of a nasogastric tube (range, 5 to 11 L; median, 7 L). Diarrhea was eventually observed in 4 equids (2 at the time of admission that had a history of diarrhea and 2 that developed diarrhea within 24 hours after admission).
A combination of CBC with fibrinogen (n = 9), serum or plasma biochemical analysis (9), venous blood gas analysis (2), and blood electrolyte analysis (2) was performed for a total of 10 equids (Table 1). Plasma lactate concentration was measured in 3 equids. Abdominal palpation per rectum was performed in 9 equids; abnormalities detected included distended small intestine (n = 2), fluid-filled large intestine (2), palpable but compressible small intestine (1), and watery feces within the rectum (1). Transabdominal ultrasonography was performed in 10 equids; abnormalities noted were distended small intestine (n = 3), fluid-filled large intestine (2), hypermotile colon (1), excess peritoneal fluid (1), hyperechoic colonic contents (1), and enlarged right and left kidneys (1). Abdominal paracentesis was attempted within 24 hours after admission in 9 equids; peritoneal fluid was obtained in 7 equids, and abnormalities were detected in 1 sample (food particles consistent with gastrointestinal rupture). Repeated abdominal paracentesis was performed in 2 equids; analysis revealed food particles consistent with gastrointestinal rupture in one equid and a mildly high peritoneal fluid nucleated cell count (6,150 cells/μL) with a protein concentration within reference range (< 2.5 g/dL) in another (ie, the 10-year-old Quarter Horse stallion).
Serum or plasma biochemical, CBC, and hematologic variables for 10 of 13 equids with persimmon phytobezoars
Variable | Median (range) | Reference range | No. of equids* |
---|---|---|---|
Hct (%) | 38.2 (37–57) | 27–43 | 2/8 |
WBCs (cells/μL) | 9,000 (5,500–14,200) | 5,600–12,100 | 4/9 |
Neutrophils (cells/μL) | 6,480 (3,685–11,700) | 2,900–8,500 | 3/9 |
Band neutrophils (cells/μL) | 0 (0–90) | 0–100 | 0/9 |
Lymphocytes (cells/μL) | 2,125 (960–4,260) | 1,160–5,100 | 2/9 |
Monocytes (cells/μL) | 134 (85–560) | 0–700 | 0/8 |
Fibrinogen (mg/dL) | 400 (200–600) | 100–400 | 3/9 |
Glucose (mg/dL) | 114 (72–428) | 62–134 | 3/10 |
Total bilirubin (mg/dL) | 2.4 (0.6–4.1) | 0–3.2 | 2/7 |
Creatinine (mg/dL) | 1.65 (0.9–4.4) | 0.4–2.2 | 2/10 |
Albumin (g/dL) | 3 (1.4–3.7) | 2.6–4.1 | 2/9 |
Globulin (g/dL) | 3.9 (2.9–7.8) | 2.6–4 | 4/9 |
Aspartate aminotransferase (U/L) | 357.5 (187–1,006) | 160–412 | 2/6 |
Creatine kinase (U/L) | 257 (143–1,828) | 60–330 | 3/9 |
γ-Glutamyltransferase (U/L) | 29 (15–145) | 6–32 | 4/9 |
Sorbitol dehydrogenase (U/L) | 14.6 (5.4–34.3) | 1–8 | 3/4 |
Alkaline phosphatase (U/L) | 221 (148–294) | 86–285 | 1/2 |
Lactate dehydrogenase (U/L) | 420 (217–623) | 112–456 | 1/2 |
BUN (mg/dL) | 18 (12–23) | 11–27 | 0/7 |
Bile acids (μmol/L) | 17.7 (7–28.9) | 0–20 | 2/4 |
Triglycerides (mg/dL) | 48 (10–48) | 6–54 | 0/3 |
Sodium (mEq/L) | 133.5 (124–139) | 128–142 | 1/8 |
Chloride (mEq/L) | 87 (83–100) | 98–109 | 4/5 |
Potassium (mEq/L) | 3.45 (3.3–4) | 2.9–4.6 | 0/8 |
Total magnesium (mg/dL) | 1.35 (1.2–1.5) | 1.4–2.3 | 0/2 |
Phosphorus (mg/dL) | 3.35 (2.8–3.9) | 1.5–4.7 | 0/2 |
Ionized calcium (mmol/L) | 0.77 (0.48–1.06) | 1.25–1.75 | 2/2 |
Total calcium (mg/dL) | 11 (10.5–11.6) | 10.2–13.4 | 0/9 |
pH | 7.312 (7.174–7.450) | 7.32–7.44 | 1/2 |
Bicarbonate (mEq/L) | 27 (14.2–33) | 24–30 | 1/4 |
Total carbon dioxide (mEq/L) | 28 (15–32.9) | 22–33 | 1/6 |
Lactate (mmol/L) | 5.9 (3.4–20) | <1.5 | 3/3 |
Number of equids with an abnormal result (ie, not within reference range)/number of equids tested for the variable.
In the 4 equids with diarrhea (2 had diarrhea at the time of admission, and 2 developed diarrhea within 24 hours after admission), results of serial bacterial cultures of fecal samples for Salmonella spp were negative. Three of these 4 equids had fecal samples tested for Clostridium perfringens and Clostridium difficile toxins and had negative results. Three of 4 equids had fecal flotations performed; 1 horse (the 10-year-old Quarter Horse stallion) had a moderate number of strongyle eggs. In the absence of overt signs of colic, diagnostic tests for fever of unknown origin or inappetence and depression were pursued in 3 equids (thoracic auscultation with and without a rebreathing bag [n = 2], thoracic ultrasonography [1], urinalysis [2], and oral examination [2]).
In the 10-year-old Quarter Horse stallion, multiple diagnostic tests were performed because of nonspecific clinical signs at the time of admission and intermittent episodes of fever (up to 40.8°C [105.5°F]) and hypoglycemia (23 to 102 mg/dL; median, 65 mg/dL), despite a fair appetite and IV administration of a 2.5% to 7.5% dextrose solution. The paired insulin-to-glucose ratio was within reference limits. Results of the d-xylose absorption test revealed blunted absorption at 0.5 to 1 hours after nasogastric administration of xylose, consistent with proximal enteral malabsorption. Results of an IV glucose tolerance test were consistent with accelerated glucose clearance with hypoglycemia (25 to 57 mg/dL) from 1 hour until discontinuation of blood sample collection at 3 hours. Histologic evaluation of a liver biopsy specimen revealed lymphoplasmacytic, neutrophilic hepatitis with portal bridging fibrosis and biliary hyperplasia.
Gastroscopy or gastroduodenoscopy was performed as a diagnostic test in 10 equids because of nonspecific findings of gastrointestinal disease (mild colic, inappetence, and weight loss). In 6 of 10 cases, historical knowledge of persimmon ingestion increased clinical suspicion of a persimmon phytobezoar obstruction. Endoscopic evidence of persimmon ingestion was visible in 8 equids, including single or multiple gastric persimmon phytobezoars (n = 7) or multiple persimmon seeds adherent to the gastric mucosa (1). In the 10-year-old Quarter Horse stallion, gastroscopy revealed multifocal (grade 2/4) gastric ulceration,1 but entry of the endoscope into the duodenum was not possible because of limited length of the endoscope (2.5 m). In another equid, gastroduodenoscopy was performed 24 hours after hospital admission (following resolution of colic signs and passage of persimmon seeds in the manure); no abnormalities were detected. Aside from the 1 horse with an enteric persimmon phytobezoar, for which the endoscope was too short to evaluate the duodenum, no other equids with enteric persimmon phytobezoars underwent gastroscopy or gastroduodenoscopy.
Two of 13 equids did not receive treatment; 1 was euthanized shortly after hospital admission because of gastrointestinal rupture, and 1 had a gastric persimmon phytobezoar, which the owner elected not to treat. Four of 13 equids were determined to have gastric persimmon phytobezoar at the time of admission on the basis of gastroscopy findings. Adjunctive treatment for colic, weight loss, and diarrhea included IV administration of isotonic or polyionic fluids (approx 2 to 10 mL/kg/h [1 to 4.5 mL/lb/h]; n = 8) with dextrose (2.5% to 7.5% solution; 2) or without dextrose (6); NSAID treatment with aspirin (15 mg/kg [6.8 mg/lb], PO, q 48 h; 1) or flunixin meglumine (0.25 to 1.1 mg/kg [0.11 to 0.5 mg/lb], q 8 h to q 24 h; 5); broad-spectrum antimicrobial treatment with potassium penicillin (22,000 to 33,000 IU/kg [10,000 to 15,000 IU/lb], IV, q 6 h; 4) and gentamicin (6.6 mg/kg [3.0 mg/lb], IV, q 24 h; 4); gastric acid suppressant treatment (4), including omeprazole (4 mg/kg [1.8 mg/lb], PO, q 24 h; 4) and cimetidine (1); sedation (3); dimethyl sulfoxide (2); enteral administration of fluids with electrolytes and dextrose or corn syrupa (2); sucralfate (1); dexamethasone (1); hypertonic saline (7.5% NaCl) solution (1); mineral oil (2); polymyxin B (1); and di-tri-octahedral smectite (1). Treatments for weight loss and diarrhea prior to diagnosis of persimmon phytobezoars included fenbendazole (n = 1) and ivermectin (1). Three equids were administered antimicrobials perioperatively, and 1 equid (the 10-year-old Quarter Horse stallion) was administered antimicrobials subsequent to findings of neutrophilic hepatitis on histologic evaluation of a liver biopsy specimen. Metronidazole was administered to 1 equid with colitis. In 1 equid, antimicrobial treatment was changed from potassium penicillin and gentamicin to enrofloxacin after surgery because of the development of an incisional infection. Treatments administered following diagnosis of persimmon phytobezoars to assist in dissolution or passage included intragastric or oral administration of carbonated colab–d soft drinks (oral administration consisted of 355 mL to 2 L of cola, q 12 h, mixed in feed, for 8 weeks [n = 1]; intermittent intragastric infusion via a nasogastric tube of 700 mL, q 12 h for 3 days [1]; intragastric constant rate infusion via a nasogastric tube consisting of 1 L of cola/h for 2 days [1]; and intragastric constant rate infusion via nasogastric tube consisting of 1 L of cola/h for 1 to 3 days, separated by 1 to 2 weeks, for 6 weeks, [1]), cellulase (2), or mineral oil (2); surgery (4); and intrapersimmon phytobezoar injections of acetylcysteine (1). Dietary modification, consisting of pelleted feed with gradually increasing amounts of hay, was recommended at the time of discharge for 8 equids.
Six equids (including the 10-year-old Quarter Horse stallion) had at least 1 episode of fever during hospitalization; 3 equids had fever at admission, and 3 developed fevers following admission. One equid was mildly febrile (38.4° to 38.7°C) from 12 to 36 hours after admission; fever resolved without treatment, and no diagnostic tests were pursued. Two equids developed fevers (38.8° to 39.3°C [101.8° to 102.8°F]) after surgery (1 equid was confirmed to have an incisional infection). Four equids had signs of colic during hospitalization; all of the episodes of colic occurred during medical management (no postoperative episodes of colic occurred during hospitalization). Serum triglyceride concentrations were measured following admission because of intermittent or persistent anorexia in 4 equids; serum triglyceride concentrations were within reference range in 3 equids (including the 10-year-old Quarter Horse stallion) and high in 1 equid (798 mg/dL). One equid developed ventral edema during hospitalization; findings on cardiac and thoracic auscultation (with and without a rebreathing bag) and serum albumin concentrations remained within reference ranges. Further diagnostic tests were not pursued because of financial constraints; the ventral edema resolved without treatment 4 days after discharge. Pharyngitis developed in 1 equid from an indwelling nasogastric tube; this resolved following treatment with throat spray consisting of dimethyl sulfoxide, glycerin, nitrofurazone, and dexamethasone. One equid had an episode of restlessness and circling without tachycardia or colic after receiving 24 L of diet colac PO in a 24-hour period; a presumptive diagnosis of caffeine toxicosis was made. Sedation and IV fluid therapy were instituted; the diet colac administration was discontinued, and signs of restlessness resolved. Ten hours later, intragastric constant rate infusion via a nasogastric tube of caffeine-free diet colad (approx 1 L/h) was begun; no further signs of agitation were observed.
A total of 5 of 13 equids with persimmon phytobezoars died; 4 of 5 equids with enteric persimmon phytobezoars died, and 1 of 8 equids with gastric persimmon phytobezoar died. Antemortem diagnosis of persimmon phytobezoars was confirmed in 9 equids. In 8 equids, gastroscopy or gastroduodenoscopy revealed the presence of 1 or multiple gastric persimmon phytobezoars. In 1 equid, an enteric persimmon phytobezoar was confirmed at surgery. One additional equid had a presumptive diagnosis of persimmon phytobezoar on the basis of historical findings of persimmon ingestion and clinical signs of colic with small intestinal distention that resolved with medical management; persimmon seeds were observed in the manure shortly after resolution of colic signs.
Gastric persimmon phytobezoars were diagnosed for 8 equids via gastroscopy or gastroduodenoscopy. Seven equids with gastric persimmon phytobezoars survived, 5 of which were managed with medical treatment or dietary modification. Medically managed gastric persimmon phytobezoars were confirmed to be resolved in 2 of 5 equids via gastroscopy; duration of treatment prior to resolution ranged from 2 days to 6 weeks. In the equid requiring 6 weeks of medical treatment, treatment consisted of intermittent intragastric constant rate infusion of diet colac via a nasogastric tube (dosage range, 1 L/h for 1 day to 1 L/h for 3 days, separated by 1 to 2 weeks). Two surviving equids with gastric persimmon phytobezoars underwent surgery after failure of prolonged medical treatment (lasting 2 and 9 weeks); at surgery, manual manipulation of the persimmon phytobezoar without gastrotomy resulted in successful fragmentation of the persimmon phytobezoar. Intraoperative or postoperative gastroscopy was performed to evaluate persimmon phytobezoar fragmentation in these 2 equids. One equid with gastric persimmon phytobezoar underwent surgery but did not survive; a gastrotomy was performed, and the equid was subsequently euthanized because of peritoneal contamination during surgery.
Enteric persimmon phytobezoars were diagnosed in 5 equids (1 diagnosis was presumptive); only 1 equid had a definitive antemortem diagnosis of enteric persimmon phytobezoar (made at surgery). One of 5 equids with enteric persimmon phytobezoars survived; in this equid, the (presumptive) enteric persimmon phytobezoar was resolved with supportive care alone. Two equids with enteric persimmon phytobezoars were euthanized because of gastrointestinal rupture; one had gastrointestinal rupture at admission, and the other had gastrointestinal rupture at readmission, 15 days after initial evaluation (5 days after initial discharge), for a complaint of mild intermittent colic. One equid with enteric (midjejunal location) persimmon phytobezoar underwent surgery, and the persimmon phytobezoar was successfully removed via enterotomy. This equid was discharged but euthanized 1 day later because of severe colic that was nonresponsive to medical management at the farm; necropsy was not performed.
The 10-year-old Quarter Horse stallion was euthanized 19 days after hospital admission because of intestinal malabsorption identified on a d-xylose test and intermittent hypoglycemia despite treatment with IV administration of dextrose solution. Necropsy revealed moderate, multifocal gastric ulceration; a duodenal persimmon phytobezoar (7 × 7 × 11 cm) and associated duodenal papillitis; moderate, diffuse lymphoplasmacytic enteritis; neutrophilic, histiocytic, and lymphoplasmacytic portal hepatitis, biliary hyperplasia and cholangitis; diffuse, moderate pancreatic acinar atrophy; and mild interstitial fibrosis and ductular hyperplasia. Necropsy specimens of the semimembranosus muscle were sent to the University of Minnesota Neuromuscular Laboratory. The histologic diagnosis based on detailed muscle evaluation was myogenic atrophy characterized by a rimmed vacuolar myopathy.
Resolution of obstruction, as assessed via gastroscopy or gastroduodenoscopy, was confirmed in 4 equids; 2 were managed medically, and 2 underwent exploratory celiotomy and external gastric massage. Time from admission to endoscopic resolution of obstruction ranged from 3 to 84 days (median, 24 days). Follow-up was performed for all surviving equids. Time from discharge to follow-up ranged from 1.5 to 7.5 years (median, 3.7 years). All equids survived > 1 year following discharge. One equid (a pregnant mare) was observed to have difficulty gaining weight immediately following discharge but was doing well at follow-up 18 months later with a normal body condition score, as reported by the owner. One equid was euthanized because of acute colic 2 years after discharge; a necropsy was not performed.
Discussion
Persimmon is a fleshy tropical fruit that may form concretions of fibers (ie, phytobezoar) within the gastrointestinal tract of humans2,3 and horses.4–7 Formation of persimmon phytobezoars occurs when the tannin monomers of the persimmon polymerize following contact with hydrochloric acid in the stomach.8 Persimmon phytobezoars can cause gastric or enteric obstruction. Furthermore, the abrasive seeds of the persimmon can become lodged within the concretions of persimmon fibers, resulting in ulceration and perforation of the gastrointestinal tract.4,9 Findings in the equids of the present report indicate that persimmon phytobezoar obstruction may result in variable clinical signs, ranging from anorexia or weight loss in equids with partial gastrointestinal obstruction to signs of severe colic in equids with complete obstruction.
Because of variable clinical signs associated with persimmon phytobezoars, obtaining a complete clinical history that includes likelihood of ingestion of persimmon is important in making a diagnosis. Ten of 13 affected equids in the present report were admitted to the hospital from September to November, whereas none were admitted from April to August. In humans, a similar seasonal association between fall persimmon harvest and persimmon phytobezoar formation has been observed.2 The persimmon tree (Diospyros virginiana) is found predominantly within the southeastern United States, with a range extending east to west from the Atlantic coast to eastern Texas and Oklahoma and extending north to south from southern Pennsylvania to Florida.10 In 5 of 9 equids of the present report for which an antemortem diagnosis of persimmon phytobezoar was obtained, history of persimmon ingestion increased the index of suspicion and resulted in early (within 24 hours after admission) gastroscopy or gastroduodenoscopy.
For equids of the present report, the most common clinical signs were colic (n = 7), weight loss (6), or anorexia (3); however, fever (3), nasogastric reflux (3), and diarrhea (2) were also noted on initial examination. Because persimmon phytobezoars cause gastroenteric obstruction, it is remarkable that colic was not a more consistent clinical sign at admission. Furthermore, in humans, the most consistent clinical signs are abdominal pain (100%) and vomiting (85% to 87%), although fever (42% to 50%) and diarrhea (19% to 20%) are also reported.2,3 In contrast to the present report on equids, weight loss is not reported for people with persimmon phytobezoars.
Common clinicopathologic abnormalities of the equids of the present report included hypochloremia (4/5 equids), high serum or plasma sorbitol dehydrogenase activity (3/4), and high serum or plasma γ-glutamyl transferase activity (4/9). Hyperglobulinemia (4/9), leukocytosis with a mature neutrophilia (3/9), and hyperfibrinogenemia (3/9) were also noted. Leukocytosis is a frequent finding in people with persimmon phytobezoars (66% to 69%); however, high serum or plasma liver enzyme activities are not observed.3,11 In the present report, hypochloremia was present in 4 of 5 equids in which serum or plasma chloride concentrations were measured. In 1 equid, hypochloremia was associated with gastrointestinal tract rupture. The 3 remaining equids with hypochloremia had enteric persimmon phytobezoars; hypochloremia may have been caused by decreased enteral absorption of chloride. In horses, high serum or plasma liver enzyme activities have been reported for multiple types of colic, including large intestine displacement or volvulus,12,13 proximal enteritis,14 or strangulating small intestinal lesions.14 In equids of the present report with persimmon phytobezoars, it is hypothesized that obstruction of the pylorus or proximal portion of the duodenum could have resulted in ileus or dysmotility, with secondary cholestasis or ascending cholangiohepatitis. Four of 6 equids with high serum or plasma liver enzyme activities had enteric obstructions, and 2 equids had large gastric persimmon phytobezoars (encompassing most of the stomach volume) that may have interfered with pyloric function.
Gastroscopy or gastroduodenoscopy was the only antemortem diagnostic test that resulted in a definitive diagnosis (in 8/9 equids) of persimmon phytobezoar obstruction. In the Quarter Horse stallion, gastroscopy revealed gastric ulceration, but the enteric persimmon phytobezoar (which was later confirmed on necropsy) was not identified because of insufficient endoscope length to reach the duodenum. One equid with a presumptive diagnosis of enteric persimmon phytobezoar had no abnormal findings on gastroduodenoscopy; however, at time of endoscopic evaluation, clinical signs had resolved. In the equids of the present report, abdominal palpation per rectum, transabdominal ultrasonography, and peritoneal fluid analysis yielded nonspecific findings of gastrointestinal disease. In addition, gastroduodenoscopy was helpful in monitoring response to treatment in 4 equids.
The Quarter Horse stallion of the present report had an enteric persimmon phytobezoar with nonspecific signs of chronic systemic inflammation, generalized muscle atrophy, and a 3-week history of weight loss despite a normal appetite that led to extensive diagnostic testing. Throughout hospitalization, this horse had intermittent hypoglycemia despite IV administration of dextrose solution. Diagnostic testing revealed proximal intestinal malabsorption and accelerated glucose clearance. Accelerated glucose clearance was hypothesized to be an adaptation to chronic intestinal malabsorption and pancreatic acinar atrophy. A rimmed vacuolar myopathy secondary to a prolonged inflammatory and catabolic state and malabsorption secondary to enteric persimmon phytobezoar obstruction was diagnosed on the basis of necropsy findings.
Findings of the present report indicate that gastric or enteric persimmon phytobezoar obstruction in equids is often amenable to medical management. Medical management, including gastric protectants, intragastric infusion of carbonated beverages or cellulase, intraphytobezoar infusion of acetylcysteine, and feeding a pelleted diet, resulted in a favorable outcome in 5 of 7 equids with gastric persimmon phytobezoars. Recently, successful dissolution of persimmon phytobezoars via oral administration or intraphytobezoar injections of colab and diet colac has been reported for humans15–19 and a horse.6 The mechanism of dissolution by colab is not well understood; however, it is likely multifactorial. It has been hypothesized that the sodium bicarbonate serves as a mucolytic20 and the carbon dioxide bubbles may disrupt the fibers of the phytobezoar,18 facilitating dissolution. Finally, the acidifying effect of cola (pH, 2.35 to 2.5) may be beneficial.21,22 In humans, 1.6 to 4 L of colab/d has been demonstrated to be effective16,17,19; extrapolating that dose to horses yields a dose of 5 to 16 L/d. For the equids of the present report, 2 of 4 treated with colab or diet colac had resolution of the persimmon phytobezoars with medical management alone. Both successfully treated equids were administered large volumes (20 to 24 L/d), whereas both equids with failed cola treatments were administered small volumes (< 4 L/d). In 1 equid with failed cola treatment, surgery was recommended but the owner elected to manage the case by feeding a pelleted diet at home; this equid was doing well at follow-up > 4 years after discharge.
In humans, gastric acid suppressants have been associated with persimmon phytobezoar formation.23 Of the 4 equids of the present report receiving gastric acid suppressants, medical dissolution failed in 2 equids, including a case of failed low-volume cola treatment. In the other 2 equids receiving gastric acid suppressants (one equid had a diagnosis of persimmon phytobezoar at necropsy and the other underwent surgery), persimmon phytobezoar dissolution was not attempted. Because of the small numbers of equids in the present report and the large variation in treatment strategies used, conclusions about the effect of gastric acid suppressants on dissolution of persimmon phytobezoars cannot be made.
Potential adverse effects of colab administration include laminitis from excess nonstructural carbohydrates and caffeine toxicosis. In 1 equid of the present report with a gastric persimmon phytobezoar, a presumptive diagnosis of caffeine toxicosis was made on the basis of clinical signs of restlessness and agitation without concurrent tachycardia or colic. This horse had received approximately 24 L of diet colac over the preceding 24 hours, which constituted a caffeine dose that has previously been associated with increased spontaneous motor activity in horses.24,25 Therefore, it is recommended that caffeinated products be used judiciously to avoid the adverse effects of caffeine. The decreased risk of adverse effects may make noncaffeinated beverages a worthwhile adjunctive treatment; however, the efficacy of these products has not yet been evaluated.
Findings of this report indicate that prolonged medical management may be required in equids with gastric persimmon phytobezoars. Successful dissolution of a gastric persimmon phytobezoar was noted after 6 weeks of intermittent treatment (1 L of diet colac/h for 1 to 3 days, separated by 1 to 2 weeks) in 1 equid. Furthermore, medical management may not always be successful. Of 3 equids undergoing prolonged medical management, 2 equids (treated for 2 and 9 weeks) yielded little progress in persimmon phytobezoar dissolution and surgery was elected. Failure of medical treatment may be because of the composition of the persimmon phytobezoar; in both equids with failed medical treatment, the persimmon phytobezoars were hard concretions that were difficult to fragment at surgery. If surgery is elected, manual manipulation of the persimmon mass to reduce its size may eliminate the need for a gastrotomy. Of the 3 equids with gastric persimmon phytobezoars in which surgery was pursued, 2 equids underwent external massage and 1 equid underwent gastrotomy. In the equid that underwent gastrotomy, peritoneal contamination occured and euthanasia was elected. Although gastrotomy has been successfully performed in horses,7,26,27 it is associated with a risk of abdominal contamination.26
On the basis of findings of the present report, gastric persimmon phytobezoars may have a better prognosis (7/8 equids survived) than enteric persimmon phytobezoars (1/5 survived). The apparent differences in prognosis between gastric and enteric persimmon phytobezoars may be due to difficulties in diagnosis and treatment of enteric persimmon phytobezoars. Furthermore, enteric persimmon phytobezoars may be more likely to cause complete obstruction (in 3/5 equids of this report) than are gastric persimmon phytobezoars (0/8 equids). Of the 5 equids with enteric persimmon phytobezoars, 3 had a definitive diagnosis at necropsy, 1 had a definitive diagnosis at surgery, and 1 had a presumptive antemortem diagnosis. Three equids with enteric persimmon phytobezoars were euthanized following confirmed (2 equids) or presumptive (1) gastrointestinal rupture secondary to enteric obstruction. Compared with other causes of nonstrangulating enteric obstructions in equids, persimmon phytobezoars may have a higher risk of gastrointestinal perforation because of the presence of abrasive persimmon seeds. Therefore, if enteric persimmon phytobezoar obstruction is suspected, early surgical intervention may be prudent. Because no persimmon phytobezoar–specific dissolution treatments were used in the 2 equids with enteric persimmon phytobezoars that had an antemortem diagnosis, the efficacy of this treatment in these types of cases is not known.
In any equid in which a persimmon phytobezoar is discovered at surgery, intra- or postoperative gastroduodenoscopy is indicated to determine whether additional persimmon phytobezoars are present. Intra- or postoperative gastroscopy was performed in 2 of 3 equids of the present report; this allowed for evaluation of efficacy of persimmon phytobezoars fragmentation. Concurrent gastric and enteric persimmon phytobezoars have been reported for people2,28 and a horse.7 For humans, reports16,29 exist of fragmented gastric persimmon phytobezoars (resulting from endoscopic fragmentation or partial dissolution) causing subsequent intestinal obstruction. Therefore, any equid undergoing treatment for persimmon phytobezoars should be closely monitored for signs of colic that could indicate consequent obstruction by a fragment of the persimmon phytobezoar.
In conclusion, gastric or enteric persimmon phytobezoar is an uncommon but potentially fatal condition of equids for which the most common clinical signs are colic, weight loss, or diarrhea. Diagnosis is often supported by historical findings and confirmed by gastroduodenoscopy. Findings of the present report indicate that equids with gastric persimmon phytobezoar may be successfully managed with medical treatment, including feeding a pelleted diet; however, in equids with severe pain or failure of response to treatment, surgery is indicated, although gastrotomy is to be avoided if possible.
Karo, ACH Food Co, Memphis, Tenn.
Coca-Cola, The Coca-Cola Co, Atlanta, Ga.
Diet Coke, The Coca-Cola Co, Atlanta, Ga.
Caffeine-Free Diet Coke, The Coca Cola Co, Atlanta, Ga.
References
1. Andrews FM, Bernard WV, Byars D, et al. Recommendations for the diagnosis and treatment of equine gastric ulcer syndrome (EGUS). Equine Vet Educ 1999; 1:122–134.
2. Krausz MM, Moriel EZ, Ayalon A, et al. Surgical aspects of gastrointestinal persimmon phytobezoar treatment. Am J Surg 1986; 152:526–530.
3. Chisholm EM, Chung SCS, Leong HT, et al. Phytobezoar: an uncommon cause of small bowel obstruction. Ann R Coll Surg Engl 1992; 74:342–344.
4. Cummings CA, Copedge KJ, Confer AW. Equine gastric impaction, ulceration, and perforation due to persimmon (Diospyros virginiana) ingestion. J Vet Diagn Invest 1997; 9:311–313.
5. Honnas CM, Schumacher J. Primary gastric impaction in a pony. J Am Vet Med Assoc 1985; 187:501–502.
6. Rodriguez-Hurtado I, Stewart A, Pellegrini-Masini A. Successful treatment for a gastric persimmon bezoar in a pony using nasogastric lavage with a carbonated cola soft drink. Equine Vet Educ 2007; 11:571–574.
7. Kellam LL, Johnson PJ, Kramer J, et al. Gastric impaction and obstruction of the small intestine associated with persimmon phytobezoar in a horse. J Am Vet Med Assoc 2000; 216:1279–1281.
8. Holloway WD, Lee SP, Nicholson GI. The composition and dissolution of phytobezoars. Arch Pathol Lab Med 1980; 104:159–161.
9. Gaya J, Barranco L, Llompart, et al. Persimmon bezoars: a successful combined therapy. Gastrointest Endosc 2002; 55:581–583.
10. USDA Forest Service Forest Products Laboratory. Common persimmon!. Available at: www.fpl.fs.fed.us/documnts/usda/amwood/250commo.pdf. Accessed Mar 21, 2010.
11. Moriel EZ, Ayalon A, Ahmed E, et al. An unusually high incidence of gastrointestinal obstruction by persimmon bezoars in Israeli patients after ulcer surgery. Gastroenterology 1983; 84:752–755
12. Gardner RB, Nydam DV, Mohammed HO, et al. Serum gamma glutamyl transferase activity in horses with right or left dorsal displacements of the large colon. J Vet Intern Med 2005; 19:761–764.
13. Southwood LL, Bergslien K, Jacobi A, et al. Large colon displacement and volvulus in horses: 495 cases (1987–1999)!, in Proceedings. 7th Int Equine Colic Res Symp 2002;32–33.
14. Davis JL, Blikslager AT, Catto K, et al. A retrospective analysis of hepatic injury in horses with proximal enteritis (1984–2002). J Vet Intern Med 2003; 17:896–901.
15. Chung YW, Han DS, Park YK, et al. Huge gastric diospyrobezoars successfully treated by oral intake and endoscopic injection of Coca-Cola. Dig Liver Dis 2006; 38:515–517.
16. Ha SS, Lee HS, Jung MK, et al. Acute intestinal obstruction caused by a persimmon phytobezoar after dissolution therapy with Coca-Cola. Korean J Intern Med 2007; 22:300–303.
17. Kato H, Nakamura M, Orito E, et al. The first report of successful nasogastric Coca-Cola lavage treatment for bitter persimmon phytobezoars in Japan. Am J Gastroenterol 2003; 98:1662–1663.
18. Ladas SD, Triantafyllou K, Tzathas C, et al. Gastric phytobezoars may be treated by nasogastric Coca-Cola lavage. Eur J Gastroenterol Hepatol 2002; 14:801–803.
19. Sechopoulos P, Robotis JF, Rokkas T. Gastric bezoar treated endoscopically with a carbonated beverage: case report. Gastrointest Endosc 2004; 60:662–664.
20. Sanderson I, Ibberson O, Fish EB. Gastric phytobezoar following gastrectomy. Can Med Assoc J 1971; 104:1115–1119.
21. Feldman M, Barnett C. Relationships between the acidity and osmolality of popular beverages and reported post-prandial heartburn. Gastroenterology 1995; 108:125–131.
22. McCloy RF, Greenburg GR, Baron JH. Duodenal pH in health and duodenal ulcer disease: effect of a meal, Coca-Cola, smoking, and cimetidine. Gut 1984; 25:386–392.
23. Nichols TW Jr. Phytobezoar formation: a new complication of cimetidine therapy. Ann Intern Med 1981; 95:70.
24. Vickroy TW, Chang SK, Chou CC. Caffeine-induced hyperactivity in the horse: comparisons of drug and metabolite concentrations in blood and cerebrospinal fluid. J Vet Pharmacol Ther 2008; 31:156–166.
25. Greene EW, Woods WE, Tobin T. Pharmacology, pharmacokinetics, and behavioral effects of caffeine in horses. Am J Vet Res 1983; 44:57–63.
26. Orsini JA, Dikes N, Ruggles A, et al. Use of gastrotomy to relieve esophageal obstruction in a horse. J Am Vet Med Assoc 1991; 198:295–296.
27. Owen R, Jagger DW, Jagger F. Two cases of equine primary gastric impaction. Vet Rec 1987; 121:102–105.
28. Vellar DJ, Vellar ID, Pucius R, et al. Phytobezoars—an overlooked cause of small bowel obstruction following vagotomy and drainage operations for duodenal ulcer. Aust N Z J Surg 1986; 56:635–638.
29. Kilam SK, Cohen MM. Small-bowel obstruction after conservative treatment of gastric bezoar. Can J Surg 1986; 29:369–371.