Transient Fanconi syndrome in Quarter horses

Abstract

Two Quarter horses with weight loss had glucosuria, euglycemia, and a mild metabolic acidosis suggesting a proximal renal tubular defect. Further testing revealed transient generalized aminoaciduria, lactic aciduria, and glucosuria, indicating Fanconi syndrome. Both horses recovered with supportive therapy. This is the first report of acquired Fanconi syndrome in horses.

In horses, 2 types of renal tubular acidoses (RTA) have been described and well-documented (1–10): type I (distal RTA), resulting from impaired excretion of H⁺ ions by the distal tubules, and type II (proximal RTA), resulting from the inability of the proximal renal tubular cells to reabsorb bicarbonate. When generalized impairment of the proximal renal tubule occurs, reabsorption of not only bicarbonate but also glucose, electrolytes, amino acids, and organic acids does not occur and the condition is termed Fanconi syndrome (11). While type II RTA and Fanconi syndrome both result in the inability of the proximal renal tubule to reabsorb bicarbonate, the hallmark metabolic status of RTA (hyperchloremic metabolic acidosis) is not a consistent feature in Fanconi syndrome. In fact, a hypochloremic metabolic alkalosis has been reported (12), although more commonly a metabolic acidosis is present with a decreased (13), normal (14–16), or increased (17,18) serum chloride concentration.

Fanconi syndrome has been well-described in humans (11,19,20) and in dogs, most notably in the Basenji breed (21–23). The syndrome can be inherited, as in Basenjis, due to an autosomal recessive trait resulting from deletion of the last exon of FAN1(24), or can occur secondary to ingestion of Chinese jerky treats, a number of drugs, toxins, heavy metals, or diseases such as copper hepatopathy and hypoparathyroidism in dogs and multiple myeloma and Wilson’s disease in humans (11,14,18,20,21,23,25–32). In dogs, the most common presenting clinical signs include weight loss, weakness, dehydration, polyuria, polydipsia, and renal failure (21). While Fanconi syndrome has not been confirmed in horses, the syndrome was suspected in 1 horse with glucosuria and proteinuria after experimental intoxication with mercuric chloride, although aminoaciduria and lactic aciduria were not described (33). To the knowledge of the authors, this report presents the first confirmed cases of acquired Fanconi syndrome in the horse.

Case descriptions

Case 1

A 13-year-old, 475 kg, Quarter horse gelding was presented with a 1-day history of abdominal discomfort, anorexia, and a right forelimb heel bulb laceration that had been treated with trimethoprim-sulfamethoxazole. No other medications were administered prior to referral by the veterinarian. Weight loss over the previous month was also noted. The horse was grazing on wheat pasture as its main forage source.

On admission, the horse was thin (body condition score 3/9) and lethargic with normal vital parameters. Mucous membranes were pink and tacky with a capillary refill time of 2 s. The horse was estimated to be 5% dehydrated. Abdominal auscultation revealed decreased borborygmi. A pelvic flexure impaction was palpable upon rectal examination. A small amount of peritoneal fluid was apparent upon abdominal ultrasonographic examination; there were no other clinical findings. An abdominocentesis was performed and cytology of the fluid was normal (total protein 17 g/L; white blood cell count < 1.0 × 10⁹/L).

Routine hematology revealed a mild anemia [hematocrit 30%; reference range (RR): 32% to 50%] and leukocytosis (14.6 × 10⁹/L; RR: 6 to 12 × 10⁹/L) with neutrophilia (12.8 × 10⁹/L; RR: 2.5 to 7.5 × 10⁹/L) and lymphopenia (0.6 × 10⁹/L; RR: 1.5 to 7.7 × 10⁹/L). A serum chemistry analysis showed a decreased bicarbonate concentration (18 mmol/L; RR: 26 to 36 mmol/L), an elevated serum creatinine (194 μmol/L; RR: 71 to 159 μmol/L), a mild hypoalbuminemia (26 g/L; RR: 27 to 37 g/L), a mild hyperglobulinemia (49 g/L; RR: 25 to 47 g/L), and a mild hyponatremia (132 mmol/L; RR: 135 to 144 mmol/L) with normal potassium (3.5 mmol/L; RR: 2.2 to 4.8 mmol/L), calcium (2.9 mmol/L; RR: 2.9 to 3.7 mmol/L), chloride (103 mmol/L; RR: 95 to 105 mmol/L), phosphorus (0.81 mmol/L; RR: 0.42 to 1.65 mmol/L), anion gap (16 mmol/L; RR: 10 to 17 mmol/L), and glucose (5.77 mmol/L; RR: 3.89 to 6.55 mmol/L) value. A free catch urine sample, prior to initiation of fluid therapy, was evaluated by reagent strip analysis and revealed a urine specific gravity of 1.031, marked glucosuria (~111 mmol/L), mild proteinuria (~1.0 g/L), and a pH of 8.0.

Initial treatment consisted of 3 boluses of enteral fluids (7 L, q1h) and intravenous lactated Ringer’s solution, 120 mL/kg body weight (BW) per day, with potassium chloride supplementation (1.7 mEq/kg BW per day) to resolve the impaction, correct the dehydration, and prevent hypokalemia while feed was withheld. Oxytetracycline (Liquamycin LA-200; Pfizer, New York, New York, USA), 6.6 mg/kg BW, IV, q12h, and flunixin meglumine (Banamine; Merck, Summit, New Jersey, USA), 1.1 mg/kg BW, IV, once, were administered for the laceration. The probiotic Saccharomyces boulardii (Florastor; Biocodex, San Bruno, California, USA), 1 mg/kg BW, PO, q12h, was administered due to antibiotic administration and risk of diarrhea.

The following morning, the horse had passed large amounts of feces and the large colon impaction was no longer palpable on rectal examination; therefore, the intravenous fluid therapy was discontinued. A possible explanation for the profound glucosuria with euglycemia was investigated. A venous blood gas analysis revealed a hyperchloremic metabolic acidosis (pH 7.29, pCO₂ 36.7 mmHg, bicarbonate 18 mmol/L, chloride 110.5 mmol/L) despite 12 h of intravenous fluid therapy. The best explanation for the decreased serum bicarbonate concentration and profound glucosuria with euglycemia was a unique generalized defect in proximal renal tubular reabsorption such as Fanconi syndrome.

A complete urinalysis was performed on the refrigerated urine sample collected at admission, which confirmed the severe glucosuria (~55 mmol/L) with a Clinitest, and mild proteinuria (~0.75 g/L) with the sulfasalicylic acid test. Sediment examination revealed rare hyaline casts, rare leukocytes, few squamous epithelial cells, and no red blood cells or bacteria. The protein-to-creatinine ratio was increased (2.8; RR: 0.6 to 0.47) (34) with abnormal fractional excretion of phosphorus (9.8%; RR: 0.023% to 2.77%) and potassium (45%; RR: 1.0% to 42.7%) and normal excretion of sodium (0.1%; RR: 0% to 2.43%) and chloride (1.6%; RR: 0.012% to 3.47%) (35). A semi-quantitative Fanconi screening test on the same urine, performed at the Metabolic Genetics Laboratory of the University of Pennsylvania (PennGen), revealed severe generalized aminoaciduria, moderate lactic aciduria, and glucosuria compared to simultaneously run and historical controls from adult horses determined to be healthy based upon clinical findings; these abnormalities are consistent with Fanconi syndrome.

Treatment to resolve the metabolic acidosis included intravenous administration of 3 L of isotonic bicarbonate (150 mEq sodium bicarbonate/L) over 4 h followed by oral sodium bicarbonate (200 mEq, PO, q6h) and potassium chloride (150 mEq, PO, q6h) supplementation to correct the metabolic acidosis and prevent hypokalemia. Repeat venous blood gas analyses guided further oral supplementation over the next 3 d. Initially, the sodium bicarbonate administration was decreased to 170 mEq, PO, q8h, and potassium chloride to 75 mEq, PO, q8h on day 3 of hospitalization and finally the frequency of administration was decreased to q12h until discontinuation on day 4 at the time of a normal acid base status (pH 7.36, total CO₂ 33 mmol/L, chloride 104 mmol/L). The leukocytosis resolved, although a mild elevation in serum globulin concentration (50 g/L) remained. The horse was transitioned to oral doxycycline (doxycycline; West-ward Pharmaceutical, Eaton Town, New Jersey, USA) at 10 mg/kg BW, PO, q12h until appropriate granulation tissue covered the laceration and the globulin fraction decreased.

The horse was discharged after 1 wk of hospitalization. Urinalysis at that time was negative for glucose and protein; a Fanconi screen was not repeated. One week later, a serum chemistry analysis revealed normal serum creatinine, bicarbonate, and globulin concentrations; therefore, the antibiotics were discontinued. Two months after discharge, the horse had markedly improved body condition and was back to his regular exercise regimen. Five months later, the serum bicarbonate (31 mmol/L) and creatinine (150 μmol/L) were still normal. The underlying cause of the Fanconi syndrome was not determined. The horse did not return to the same pasture and continues to do well.

Case 2

A 27-year-old, 395 kg, Quarter horse mare was presented with a 10-day history of poor appetite and chronic weight loss (27 kg since the previous year). The horse had been taken to the hospital annually for the past 3 y for routine dental floatation.

On admission, the horse had a body condition score of 3/9 and normal vital parameters. A CBC was within normal limits. Serum chemistry analysis revealed azotemia with elevated blood urea nitrogen (9.6 mmol/L; RR: 3.2 to 7.9 mmol/L) and creatinine concentration (327 μmol/L), mild hypophosphatemia (0.36 mmol/L), mild hyponatremia (134 mmol/L), decreased bicarbonate concentration (21 mmol/L), and a normal anion gap (15 mmol/L), and normal serum glucose (6.3 mmol/L), potassium (3.7 mmol/L), chloride (104 mmol/L), and calcium (3.3 mmol/L) concentrations. A urine sample was collected by catheterization, without sedation, prior to initiating fluid therapy. The urine specific gravity was 1.016 and reagent strip analysis revealed moderate glucosuria (~28 mmol/L), a positive heme reaction (2+), and a pH of 8.5. The kidneys appeared normal by ultrasonographic examination.

Due to the apparent kidney injury, intravenous fluid therapy was initiated overnight with lactated Ringer’s solution (120 mL/kg BW per day) with potassium chloride (2.2 mEq/kg BW per day) and dextrose (Vedco, St. Joseph, Missouri, USA) supplementation (3 mg/kg BW per day) due to prolonged anorexia. The cause for glucosuria with euglycemia was investigated and in the presence of a mild decrease in the serum bicarbonate concentration, Fanconi syndrome was suspected.

The next morning, venous blood gas analysis revealed a hyperchloremic metabolic acidosis (pH 7.30, pCO₂ 37.2 mmHg, bicarbonate 17 mmol/L, chloride 112 mmol/L). Analysis of the urine sample obtained at admission, confirmed glucosuria (~14 mmol/L) with a Clinitest. No red blood cells, casts, epithelial cells, or bacteria were seen with only occasional leukocytes present on microscopic examination of the sediment. The urine protein-to-creatinine ratio was normal. Fractional excretion of sodium (0.36%), chloride (0.42%), and phosphorus (0.48%) was normal with an elevation in potassium excretion (56%). The Fanconi screening test of urine at PennGen showed mild aminoaciduria, lactic aciduria, and glucosuria, compared to simultaneous run and historical controls from adult horses determined to be healthy based on clinical findings, which is consistent with Fanconi syndrome.

To correct the metabolic acidosis, 3 L of isotonic bicarbonate were administered over 4 h (150 mEq/L sodium bicarbonate) followed by oral sodium bicarbonate supplementation (160 mEq, PO, q4h). Repeat venous blood gas evaluations, performed every 12 to 24 h during hospitalization, revealed a persistent metabolic acidosis (day 4: pH 7.3, pCO₂ 41 mmHg, bicarbonate 20.2 mmol/L) despite intensive sodium bicarbonate supplementation. The horse remained azotemic with a serum creatinine concentration of 354 μmol/L 4 days after presentation. Although a renal biopsy for histopathology and further treatment were offered, the owner elected to take the horse home.

The horse was supplemented with oral bicarbonate (470 mEq, PO, q12h) for 4 wk, at which time the glucosuria and proteinuria had resolved and the blood bicarbonate concentration had normalized (31 mmol/L) with a high normal creatinine (159 μmol/L). Ten months later, the horse had gained 51 kg and the urinalysis was unremarkable. The urine metabolic screen at that time still indicated very mild aminoaciduria without glucosuria or lactic aciduria, interpreted as resolving Fanconi syndrome. As of this writing and at 29 years of age, the mare has gained 100 kg, her body condition score is 6/9, and the Fanconi syndrome appears to have resolved (urine was negative for glucose with a normal amino acid pattern), although the mare remains in chronic renal failure (serum creatinine 230 μmol/L, urine specific gravity 1.013). No trigger for the transient Fanconi syndrome was identified.

Discussion

Hereditary and acquired Fanconi syndrome are most commonly reported in humans and dogs. The 2 adult Quarter horses reported here were presented with weight loss, glucosuria, lactic aciduria, and aminoaciduria and are the first documented cases of acquired Fanconi syndrome in horses. Contrary to the findings in previous reports on RTA in horses (2–7,10), and consistent with the definition of Fanconi syndrome, both cases reported here had glucosuria with euglycemia and defective reabsorption of amino acids by the proximal renal tubules.

Horses most commonly develop glucosuria secondary to hyperglycemia associated with pituitary pars intermedia dysfunction (PPID), stress from critical illness, or the use of sedatives (36–40). Neither horse had clinical evidence of PPID nor were sedatives administered prior to urine collection. It is important to note that the urinary reagent strip analyses providing evidence of moderate to marked glucosuria at presentation were performed immediately after urine collection and the urinalyses with examination of the sediment were performed the next day, possibly leading to decreased glucose concentrations from metabolism by white blood cells and epithelial cells during storage.

Specific causes of glucosuria with euglycemia in the horse are not well-described in the literature. However, in other species including dogs, isolated glucosuria (e.g., elkhounds) and glucosuria with other tubular defects (Fanconi syndrome) have been well-described (41). The transient glucosuria with a mild metabolic acidosis and a normal serum glucose and anion gap, suggested the presence of transient Fanconi syndrome in the horses in this report. Therefore, treatment was initiated and diagnostic investigation into Fanconi syndrome was pursued.

A metabolic screening test of urine is available for small animals, which evaluates urine semi-quantitatively for the presence of carbohydrates (including glucose), all amino acids (which are normally resorbed in the proximal tubules), and organic acids (23). The test assesses free amino acids in urine which is different from the determination of proteinuria. Animals with Fanconi syndrome have generalized and sometimes massive aminoaciduria with or without proteinuria and thus aminoaciduria cannot be diagnosed by routine urinalysis. Aminoaciduria is frequently observed without proteinuria but with other tubular defects resulting in increased urinary concentrations of glucose and bicarbonate, as observed in the horses in this report. While the test is set up for the detection of inborn errors of metabolism, such as Fanconi syndrome in Basenjis or cystinuria (a specific proximal tubular defect), it can also be used for acquired forms of Fanconi syndrome. Recently the test was used to evaluate a number of small breed dogs that developed Fanconi syndrome following the ingestion of Chinese jerky treats and dogs with copper hepatopathies (14,32).

The test has not been validated for use in horses, but the results from these horses were compared to the results from over 50 other horses determined to be healthy or with various medical conditions including cases of renal disease. Each urine sample was also simultaneously analyzed with urine from a healthy horse for comparison. The urinary metabolic pattern from the other horses evaluated showed the expected hippurate excretion, but no lactic acids or glucose and only distinct amino acids in very low concentration as they are normally all efficiently reabsorbed. Also, dogs with acute and chronic renal injury do not show aminoaciduria or glucosuria; therefore, the presence of lactic aciduria, aminoaciduria, and glucosuria, in light of normal to low blood glucose levels is indicative of Fanconi syndrome. Fanconi syndrome in dogs varies in severity, with some showing glucosuria, aminoaciduria, and lactic aciduria with severe metabolic derangements and others showing only aminoaciduria and mild acid-base abnormalities. The horses presented here appear to have had mild to moderate and transient Fanconi syndrome based on the metabolic status and relatively rapid response to treatment. While the urine metabolic findings present in these 2 horses have never been observed in previously tested horses, continued urine screening of horses with other diseases, including those that cause glucosuria such as pars pituitary intermedia dysfunction, may lead to validation of this test in horses.

The inciting cause leading to Fanconi syndrome in either horse was not determined and the condition resolved within days to a month after the initial diagnosis. As the Fanconi syndrome in these horses occurred in advanced age and was transient, an acquired cause appears likely. Causes of acquired Fanconi syndrome in humans include heavy metal intoxication (30,31), exposure to antiviral drugs such as Tenofovir (20), outdated tetracyclines (26), and diseases such as Wilson’s disease (25) and multiple myeloma (27). In dogs, Fanconi syndrome has been associated with gentamicin toxicity (28), amoxicillin (29), Chinese jerky treats (23,32), hypoparathyroidism (18), and copper-associated hepatopathy (14,42). The only therapy administered to the first horse prior to presentation was trimethoprim-sulfamethoxazole and there was no history of drugs being administered to the second horse. Reports in humans exist regarding the use of trimethoprim-sulfamethoxazole and the development of RTA, although there is no evidence for the development of Fanconi syndrome (43,44).

The development of azotemia and chronic renal failure with Fanconi syndrome has been documented in humans (45,46) and dogs (16,22). Yearley et al. (22) found that among 17 Basenjis diagnosed with Fanconi syndrome with long-term follow-up available, 8 developed persistent azotemia and most were euthanized or died as a consequence of renal failure. Follow-up serum biochemistry analyses were available in both horses reported here in addition to repeat urinalyses in the second horse. Only 1 horse, however, had evidence of chronic renal failure which persisted despite resolution of the Fanconi syndrome. While likely, it is unknown whether the same insult caused the renal failure and the Fanconi syndrome.

Fanconi syndrome in humans and dogs is usually hereditary and progressive, although there are reports of transient disease (15,17,18,23,32) best exemplified by the ingestion of Chinese jerky treats. The dogs reported to have transient disease had no evidence of relapse after removal of the inciting cause (Chinese jerky treats). Similarly, neither horse in this report had evidence of a relapse after the initial treatment regimen.

In conclusion, Fanconi syndrome should be considered in horses with glucosuria and euglycemia. The Fanconi test that was originally developed for urinary metabolic genetic screening for inborn errors of metabolism can be used in horses as the metabolites are identical.