Distal Renal Tubular Acidosis Associated with Concurrent Leptospirosis in a Dog

Introduction

Renal tubular acidosis (RTA) is rarely reported in veterinary medicine compared with its occurrence in humans. There are reports of distal RTA in dogs, cats, horses, and cows.^1–3 RTA is classified by a normal anion gap, hyperchloremic metabolic acidosis occurring as a result of either decreased bicarbonate (HCO₃) reabsorption (i.e., proximal RTA) or defective acid excretion (distal RTA) in the presence of a normal glomerular filtration rate.^3–7 Lethargy is the most frequent historical finding observed in dogs.⁸ In humans, other findings include muscle weakness, vomiting, inappetence, weight loss, stunted growth, constipation, polyuria, polydipsia, and paralysis.^3,8 Osteomalacia and urolithiasis are also reported consequences of distal RTA and chronic metabolic acidosis,and progression of nephrocalcinosis may lead to chronic renal failure.^3,5

Leptospirosis is a significant bacterial disease in humans and dogs.⁹ Over 250 pathogenic serovars are described and are classified into antigenically related serogroups.⁹ Disease in dogs is primarily caused by Leptospira interrogans and Leptospira kirschneri.¹⁰ The most common serovars thought to cause disease in dogs in the United States include icterohemorrhagiae, canicola, grippotyphosa, pomona, bratislava, and autumnalis.⁹ Renal tubular infection by leptospires is associated with acute interstitial nephritis and tubular dysfunction. Leptospires have been demonstrated to colonize the renal tubules and multiply in the apical aspect of the renal tubular epithelial cells of a diverse array of mammalian reservoirs.^9,11,12 Leptospirosis has been associated with renal tubular defects in humans.² Despite these associations, distal RTA has never been reported in a dog with an active infection due to leptospirosis. The purpose of this case report is to describe the first case of distal RTA in a dog with concurrent active infection with leptospirosis.

Case Report

A 9 yr old spayed female boxer was presented for evaluation of vomiting, lethargy, anorexia, and weight loss. Clinical signs began approximately 1 wk prior to referral. The dog had been obtained by the owner as a puppy and had lived in northwest Ohio for its entire life. The patient had been vaccinated against distemper, adenovirus type 2, parvovirus, parainfluenza, leptospirosis, rabies, and Bordetella 8 mo prior to presentation. The leptospirosis vaccine included Leptospira serovars grippotyphosa, pomona, canicola, and icterohemorrhagiae. The dog was receiving a combination of ivermectin and pyrantel pamoate^a and fipronil and s-methoprene^b for prevention of heartworm and flea/tick-borne diseases, respectively. Previous medical history included a diagnosis of hypothyroidism, and the patient was receiving levothyroxine (.5 mg per os [PO] q 12 hr). The dog was presented to the primary veterinarian 2 days after clinical signs began. Diagnostics included a complete blood count, serum biochemical analysis, and urinalysis. Abnormalities included elevated lipase (60.04 μkat/L; reference range, 3.34–30.06 μkat/L) and chloride ([Cl]; 130 mmol/L; reference range, 109–122 mmol/L) and a decrease in the total white blood cell count (4.74 × 10⁹/L; reference range, 5.5–16.9 × 10⁹/L). Individual cell lines were within normal limits. Urinalysis revealed isosthenuria (specific gravity was 1.010) and a pH of 8. Urine sediment evaluation did not reveal abnormalities. The patient was hospitalized and placed on IV fluids (type and rate unavailable for review) with potassium (K) Cl supplementation, IV ampicillin, and subcutaneous maropitant^c (doses not available for review). The dog was discharged 2 days later on metoclopramide but continued to vomit. The dog was presented 2 days later, and an upper gastrointestinal barium series was performed to rule out intestinal obstruction. It was reported that there was no evidence of obstruction, but decreased gastrointestinal motility was suspected due to delayed gastric emptying. The radiographs were not available for review. The patient was then referred for evaluation of either a suspected gastrointestinal foreign body or gastric outflow obstruction.

Upon referral, the dog was quiet, alert, and responsive. Vital parameters were within normal limits. The dog weighed 21.8 kg, with a body condition score of 4/9. Physical examination revealed a prolonged capillary refill time (> 3 sec) and moderate dehydration (6–8%). Initial laboratory evaluation consisted of a complete blood count, serum biochemical analysis, canine pancreatic lipase, urinalysis, and urine culture. Hemoconcentration was noted with elevations in hematocrit (57%; reference range, 37–55%) and hemoglobin (200 g/L; reference range, 120–180 g/L). Elevated blood urea nitrogen ([BUN], 11.78 mmol/L; reference range, 2.50–9/64 mmol/L), Cl (150 mmol/L; reference range, 105–115 mmol/L), and Na (162 mmol/L; reference range, 141–156 mmol/L), and decreased HCO₃ (5 mmol/L; reference range, 17–24 mmol/L) were also noted. The anion gap was within normal limits (12 mmol/L; reference range, 12–24 mmol/L). Pancreatic lipase was mildly increased (234 μg/L; reference range, ≤ 200 μg/L). Urinalysis obtained via cystocentesis revealed inappropriate urine concentration (specific gravity, 1.017), proteinuria (3+), and a pH of 7. Urine sediment evaluation detected 6–10 epithelial cells/high-power field, and a urine culture was negative. A urine protein/creatinine ratio was performed and was elevated (1.3; reference range, < 0.2). Venous blood gas analysis revealed a marked metabolic acidosis (pH was 6.893, partial pressure of CO₂ was 22.3 mmHg, base excess in extracellular fluid was −29, HCO₃ was 4.3 mmol/L, and total CO₂ was < 5 mmol/L). Thoracic radiographs and abdominal ultrasound findings were within normal limits. The dog was placed in the intensive care unit, and a balanced electrolyte solution^d was administered IV at 66 mL/kg/day with 20 mEq KCl/L with 5% dextrose in water at 66 mL/kg/day to correct for dehydration and free water deficit. Initial treatments included ampicillin Na/sulbactam Na^e (28 mg/kg IV q 8 hr), famotidine^f (1 mg/kg IV q 24 hr), and a metoclopramide^g continuous rate infusion (2 mg/kg/day IV). Antibiotics were administered while urine culture was pending to treat for potential pyelonephritis. Due to the severe acidosis, a 4 hr constant rate infusion of 90 mEq of NaHCO₃ was initiated with re-evaluation of venous blood gases at the end of each aliquot until the venous pH was > 7.2. Given a normal anion gap, hyperchloremic metabolic acidosis with inappropriately alkaline urine, and presumed normal glomerular filtration rate, distal RTA was strongly suspected.

Once vomiting and regurgitation were controlled, oral HCO₃ therapy^h was started (11 mEq PO q 8 hr) with continued blood gas monitoring. Due to a decline in venous pH (7.232–7.086), HCO₃ was increased (20 mEq PO q 6 hr). On day 2, the dog developed hypokalemia (2.8 mmol/L), and K gluconate was initiated (10 mEq PO q 12 hr). Additional treatments included mirtazapineⁱ (7.5 mg PO q 24 hr) and maropitant (1 mg/kg subcuntaneously q 24 hr). The dog was hospitalized for an additional 4 days on the therapy described above. During that time, continued improvement in appetite was noted. At the time of discharge, electrolyte analysis revealed an improvement in Cl (139 mmol/L; reference range, 105–115 mmol/L), Na (157 mmol/L; reference range, 141–156 mmol/L), K (3.3 mmol/L; reference range, 3.5–5.0 mmol/L), and venous blood gas analysis (pH was 7.209, partial pressure of CO₂ was 26.6 mm Hg, base excess in extracellular fluid was −17, HCO₃ was 10.6 mmol/L, and total CO₂ was 11 mmol/L). The anion gap remained normal (11 mmol/L; reference range, 8–24 mmol/L)^j. The dog was discharged on amoxicillin trihydrate/clavulanate^k (11.4 mg/kg PO q 12 hr) for 2 wk, maropitant (2.7 mg/kg PO q 24 hr) as needed for vomiting, mirtazapine (7.5 mg PO q 24 hr) for 2 wk, metoclopramide (0.22 mg/kg PO q 8 hr) for 1 wk, and famotidine (0.9 mg/kg PO q 24 hr) for 1 mo. Maintenance alkali therapy consisted of K gluconate^l (10 mEq PO q 12 hr) and NaHCO₃ (25 mEq PO q 8 hr).

The dog was re-evaluated 5 days later, and the owner reported a mildly decreased appetite from the time of discharge with a stable weight and marked polyuria and polydipsia. Diagnostics performed included a renal chemistry (albumin, globulin, BUN, creatinine, cholesterol, Ca, phosphorus, HCO₃, Cl, K, and Na), thyroxine, venous blood gas, urinalysis, and urine culture. There was an improvement in venous pH (7.26), Cl (133 mmol/L), Na (145 mmol/L), and K (4.3 mmol/L). Renal chemistry abnormalities (BUN, Na, K, Cl, and HCO₃) were similar to those reported from venous blood gas analysis, and all other values were within normal limits. Urinalysis revealed hyposthenuria (specific gravity was 1.004), and the urine culture was negative. Leptospirosis microscopic agglutination testing (MAT), which included serovars bratislava, canicola, grippotyphosa, icterohemorrhagiae, pomona, and autumnalis, and an adrenocorticotropic hormone (ACTH) stimulation test were also performed given the marked hyposthenuria. Cortisol results were within normal limits (preACTH was 52.42 nmol/L; reference range, 55.18–165.53 nmol/L and postACTH was 201.39 nmol/L; reference range, 5–15 μg/dL). Serologic evaluation for leptospirosis revealed positive titers for grippotyphosa (1:400), canicola (1:200), icterohemorrhagiae (1:200), pomona (1:200), and autumnalis (1:100). Thyroid testing revealed a decreased total thyroxine (< 6.84 nmol/L; reference range, 17.09–68.38 nmol/L) and increased canine thyroid-stimulating hormone (26.1 mIU/L; reference range, 4.6–25.6 mIU/L). Results of the leptospirosis testing were interpreted as most consistent with a vaccinal response. As convalescent titers had not yet been performed, amoxicillin trihydrate/clavulanate K was discontinued and doxycycline^m (4.5 mg/kg PO q 12 hr) was initiated, and convalescent titers were recommended in 2 wk. Levothyroxine was subsequently restarted at the previously described dose. Given the results of diagnostic testing combined with the report of a sudden and unexplained increase in water consumption and hyposthenuria after initiation of therapy, a modified water deprivation test was performed.¹³ Initial urine specific gravity revealed hyposthenuria (specific gravity was 1.006). Blood samples were drawn q 6 hr to examine BUN, electrolytes, hematocrit, hemoglobin, osmolality, and acid/base status. The patient was weighed and urine specific gravity was monitored q 4 hr. Aside from a persistent metabolic acidosis (pH was 7.26–7.34), all parameters remained within normal limits throughout the study. After 24 hr, the urine specific gravity showed adequate urine concentration (1.030) with a 5% loss in body weight consistent with psychogenic polydipsia. The dog was discharged with instructions to limit water consumption to 40–60 mL/kg/day.

Re-evaluation was performed 2 wk later, and the owner reported normal attitude, appetite, and water consumption. Venous blood gas revealed an improvement in pH (7.332), Cl (136 mmol/L), Na (146 mmol/L), and K (3.6 mmol/L). Convalescent leptospirosis serology revealed an increased titer to grippotyphosa (1:800) and autumnalis (1:3,200). All other titers were either decreased or equal compared with previous measurement. Maropitant, mirtazapine, and metoclopramide were discontinued, and doxycycline was continued for an additional 2 wk. Alkali therapy was continued as previously described because the distal RTA was not expected to resolve. In the following month, the patient was treated by the primary veterinarian for gastritis with 2 wk of metronidazole (10.8 mg/kg PO q 12 hr) but otherwise remained free of clinical signs of distal RTA. Blood gas analysis 1 mo later revealed similar venous pH (7.26), Cl (133 mmol/L), Na (145 mmol/L), and K (4.3 mmol/L), and the owner reported a good quality of life.

Discussion

RTA is rarely reported in veterinary medicine compared with its occurrence in humans. There are reports of distal RTA in dogs, cats, horses, and cows.^1–3 RTA is classified by a normal anion gap, hyperchloremic metabolic acidosis occurring as a result of either decreased HCO₃ reabsorption (proximal RTA) or defective acid excretion (distal RTA) in the presence of a normal glomerular filtration rate.^3–7 In the human literature, RTA has been classically divided into four subtypes. The dog in this report had characteristics most similar to type I or distal RTA, which occurs secondary to a decreased fractional excretion of HCO₃ and an impaired capacity for excretion of hydrogen (H) ions in the distal tubule, producing abnormally high urine pH during systemic acidosis.^3,5 Other variants of RTA include type II or proximal RTA, which occurs secondary to a deficiency in HCO₃ reabsorption in the proximal tubule but the ability to maximally acidify the urine is retained. Type III RTA is characterized by a combination of proximal and distal RTA, usually caused by a carbonic anhydrase deficiency. That form of RTA has been proven to be a variant of type 1. Type IV RTA, hyperkalemic RTA, occurs secondary to hyporeninemic hypoaldosteronism.^{3,5,6,14–16}

Diagnosis of distal RTA in this patient was based on the presence of a severe hyperchloremic nonanion gap metabolic acidosis with a urine pH of > = 5.5–6. That type of metabolic acidosis with a urine pH > 6 is diagnostic for distal RTA in humans.¹⁷ Pyelonephritis with urease-producing bacteria can also cause similar signs but was ruled out by negative urine cultures. Severe diarrhea can lead to a normal anion gap metabolic acidosis, through loss of HCO₃ but was not present in this dog.^3,5 Chronic kidney disease is a rule out for a normal anion gap metabolic acidosis.³ The patient described in this report did present with a mildly elevated BUN, but the degree of acidosis was considered inappropriate for the level of azotemia and no evidence of chronic renal changes were noted on abdominal ultrasound. In addition, the ability to maximally acidify the urine is usually maintained in renal insufficiency.¹⁸ Prerenal azotemia is the most likely explanation as the BUN normalized with fluid therapy. Additionally, the results of the modified water deprivation test demonstrated normal urine concentrating ability, indicative of adequate renal function.

In humans, distal RTA is recognized by the inability to decrease urine pH to < 5.5 in spite of metabolic acidosis and is often accompanied by low urinary ammonia, hypocitraturia, and hypokalemia.⁵ Three mechanisms have been proposed to explain the pathogenesis of distal RTA. Classic (type I) distal RTA results from failure of the H ion-pumping adenosine triphosphatase (ATPase) pump in the basolateral membrane of α-intercalated cells.^3,6,7,19 This form results in K wasting as H cannot be pumped out of the cell and H is drawn from the cell to maintain electroneutrality.³ Another proposed mechanism is related to increased permeability of the distal tubule membrane, which is normally impermeable to H.^3,19 H then combines with hydroxide ions intended for HCO₃ production.¹⁰ That also results in increased K exchange as K is used as a substitute cation for H in exchange for Na, leading to metabolic acidosis with hypokalemia.^3,8 The final form, voltage-dependent RTA, occurs because of impaired Na reabsorption.³ That disrupts the electrical gradient that facilitates the excretion of H and K out of the cell, leading to metabolic acidosis and hyperkalemia.³ Distal RTA occurs as either a primary or secondary disorder and can be caused by autoimmune disease (e.g., systemic lupus erythematous, Sjogren syndrome in humans), drug administration (e.g., amphotericin B, tetracyclines), toxins (e.g., heavy metals), pyelonephritis, chronic hypocalcemia, and either genetic or structural disruptions of renal tubules.^1,5,6 Plasma HCO₃ concentrations tend to be lower in humans with distal RTA than those with proximal RTA.¹

Lethargy is the most frequent historical finding observed in dogs.⁸ In humans, other findings include muscle weakness, vomiting, inappetence, weight loss, stunted growth, constipation, polyuria, polydipsia, and paralysis.^3,8 Osteomalacia and urolithiasis are also reported consequences of distal RTA and chronic metabolic acidosis.³ Progression of nephrocalcinosis may lead to chronic renal failure.⁶ Those signs were not noted in the patient described in this report, likely due to the acute onset of the disease process given that osteomalacia and urolithiasis are associated with chronic disease.

Differentiation of proximal and distal RTA can be determined by a HCO₃ challenge test.^3,10 In humans, a diagnosis of distal RTA is made when three timed urine samples are obtained with a pH > 7.5 after a HCO₃ infusion sufficient to sustain an increase in serum HCO₃ concentration of 0.5–1 mEq/L/hr.^6,8 A HCO₃ challenge test was not performed in the patient described herein given that it is unnecessary with an elevated urine pH associated with systemic acidemia.⁸ An ammonium Cl loading test can also be considered to confirm diagnosis of distal RTA. To do that, urine pH is monitored before and q 1 hr for 5 hr after oral administration of 0.2 g/kg ammonium Cl. In normal dogs, urine pH should decrease to a minimum value of 5.16 by 4 hr postadministration.⁷ That test is often unnecessary as animals with RTA typically have metabolic acidemia that maximally stimulates excretion of acid by the kidneys and is only indicated in people if systemic acidemia is absent.^8,20 That test also was not performed as vomiting was a primary complaint and is commonly associated with ammonium Cl administration and it was deemed unsafe to delay treatment of the length of time it would take to perform the test.¹⁷ To characterize distal RTA as either a defect in the H ATPase or voltage-dependent, a furosemide response test can be performed.^6,10 That test is performed via the administration of 1–2 mg/kg furosemide either PO or IV.²¹ Furosemide increases luminal electronegativity by increasing cortical distal tubular Na delivery and reabsorption. Patients with a cortical H ATPase defect will have persistently alkaline urine with an increase in K secretion. Conversely, a defect within the medullary collecting tubule will produce an increase in H (i.e., urine pH < 5.5) and K. The voltage-dependent form will remain unchanged.^6,10 That test was not performed as the degree of dehydration and critical nature of the patient precluded the use of furosemide.

Treatment of distal RTA in this patient was primarily directed at identifying any underlying cause, but alkali administration was necessary to restore the acid/base balance. In veterinary medicine, initial HCO₃ doses of 1–1.5 mEq/kg/day have been recommended, but doses as high as 2–4 mEq/kg/day may be required to maintain a normal pH.^3,7 Distal RTA typically requires a lower amount of alkali therapy for management than proximal RTA. In humans with distal RTA, the amount of HCO₃ required to control acidemia is modest. Doses of 0.5–3 mEq/kg/day have been recommended as initial dosage to sustain correction of distal RTA.⁵ Conversely, in infants with proximal RTA, dosages of 5–14 mEq/kg/day may be required to completely resolve clinical signs.^22,23 K supplementation should also be considered as alkali administration may further reduce plasma K by increasing urinary loss as well as altering the distribution of K between the extracellular and intracellular fluid.^1,5,8 The dose of HCO₃ supplementation needed to control academia in the patient described in this report (4 mEq/kg/day) is still considered modest and consistent with a diagnosis of distal RTA. K gluconate supplementation was chosen in this dog not only to correct the observed K deficiency but also as an additional source of alkali in management of metabolic acidosis.⁷

Leptospirosis is a significant bacterial disease in humans and dogs.⁹ Over 250 pathogenic serovars are described and are classified into antigenically related serogroups.⁹ Disease in dogs is primarily caused by Leptospira interrogans and Leptospira kirschneri.¹⁰ The most common serovars thought to cause disease in dogs in the United States include icterohemorrhagiae, canicola, grippotyphosa, pomona, bratislava and autumnalis.⁹

Use of antibody testing for diagnosis of leptospirosis generally is based on the MAT and is the current diagnostic test of choice.⁹ A four-fold increase in convalescent titers 7–14 days after initial testing supports recent infection, although an increase in titer may be blunted by antimicrobial therapy.⁹ Titers resulting from previous vaccination, exposure, or chronic infection generally change either more slowly or not at all, and generally decline by 4 mo after vaccination.⁹ Antibodies within the same serogroup cross-react extensively and higher cross-reactive titers or “paradoxical reactions” can occur to a noninfecting serovar and are especially common in early infection.⁹ Vaccines are not thought to be cross-protective against nonvaccinal serovars.¹¹ Initial MAT testing in the patient described herein showed the strongest reaction to grippotyphosa at 1:400. Convalescent titers revealed a two-fold increase in that titer to 1:800 and a five-fold increase in autumnalis from 1:100 to 1:3,200. The patient in this report had been previously vaccinated with a four-way vaccine 8 mo previously and had been annually vaccinated for the past 3 yr. Records were unavailable prior to 2008, and the authors were unable to confirm if the dog had received initial vaccines in accordance with manufacturer recommendations. A lack of appropriate vaccination may explain the lack of vaccine efficacy in this case, and those results are considered inconsistent with a vaccinal response as titers are typically stable, absent, or declining 8 mo postvaccine. A positive titer to autumnalis seems to be a common result even in vaccinated research dogs and other dogs that have not been exposed to that serovar.¹¹ As vaccine serovars are not cross-protective, the results may be best explained by active infection with an unidentified serovar with a paradoxical reaction to autumnalis, although vaccine failure cannot be ruled out based on current testing.

Renal tubular infection by leptospirosis is associated with acute interstitial nephritis and tubular dysfunction. Leptospires have been demonstrated to colonize the renal tubules and multiply in the apical aspect of the renal tubular epithelial cells of a diverse array of mammalian reservoirs.^9,11,12 Leptospirosis has been associated with renal tubular defects in humans.¹² Polyuria and polydipsia can develop with dogs in the absence of azotemia, and those patients may be hyposthenuric.⁹ Several human studies have shown that the distal tubules showed damage before other segments, particularly the thick ascending limb of Henle’s loop, although that is less common than proximal tubular involvement.^12,24 Experimentally, leptospiral infection can cause decreased vasopressin responsiveness, suggesting a form of acquired nephrogenic diabetes insipidus.⁹ The results of a modified water deprivation test were most consistent with transient psychogenic polydipsia; however, distal RTA, leptospirosis infection, or any of the treatments received have not been associated with psychogenic polydipsia. A modified water deprivation test is not indicated in an animal with active leptospirosis infection and can be potentially harmful in a patient recovering from leptospira colonization. Initial MAT showed similar titers for all serovars and was initially interpreted as most consistent with a vaccinal response. In retrospect, it would have been prudent to await results of convalescent titers as the patient was being treated with a amoxicillin trihydrate/clavulanate, an accepted treatment of leptospirosis; however, the owner expressed extreme frustration at the amount of polyuria and polydipsia and questioned long-term quality of life if the condition was not addressed. The owner did report a sudden and dramatic reduction in water consumption after initiation of doxycycline started days prior to the modified water deprivation test, but initial urine specific gravity prior to initiation of the test showed persistent hyposthenuria. The authors felt compelled to provide an explanation and solution for the sudden overall increase in water consumption. The results of testing were most consistent with psychogenic polydipsia, but a more likely explanation for the hyposthenuria noted was transient nephrogenic diabetes insipidus secondary to active infection with leptospirosis and subsequent recovery with appropriate antibiotic therapy.

The prognosis of distal RTA is not well understood due to the paucity of case reports in veterinary medicine, but it is generally considered worse than proximal RTA due to the risk of complications such as urolithiasis and bone demineralization.^3,5 Humans with distal RTA commonly form uroliths composed of Ca phosphate; however, Ca oxalate and struvite uroliths may also occur.¹⁵ Those changes can be prevented with early and complete correction of the acidemia.¹⁸ Management of distal RTA is likely lifelong, although reports of transient RTA do exist.^3,25 Clinical signs and biochemical abnormalities in this dog were controlled on approximately 4 mEq/kg NaCHO₃ and 0.5 mEq/kg of K/day. Frequent evaluations were recommended to assess metabolic status and maintain electrolyte homeostasis.^4,6

Conclusion

Distal RTA is an extremely rare condition in veterinary medicine. To the authors’ knowledge, this is only the seventh reported case of distal RTA in the dog and the first reported case of distal RTA associated with active leptospirosis in any species.^15,20,26 Although rare, leptospirosis infection should be included on the differential list of any animal with a diagnosis of RTA. Six mo after initial diagnosis, the patient continued to remain stable with a good quality of life and no significant complications from the disease.