Correction of Hyperkalemia in Dogs with Chronic Kidney Disease Consuming Commercial Renal Therapeutic Diets by a Potassium-Reduced Home-Prepared Diet
All dogs were treated at the William R. Pritchard Veterinary Medical Teaching Hospital, University of California, Davis, One Shields Avenue, Davis, CA. Segev is presently affiliated with School of Veterinary Medicine, the Hebrew University of Jerusalem, Israel. Weeth is presently affiliated with Clinical Nutrition Department, Red Bank Veterinary Hospital, Tinton Falls, NJ.
Corresponding author: Gilad Segev, School of Veterinary Medicine, the Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel; e-mail: firstname.lastname@example.org.
Background: Hyperkalemia occurs in dogs with chronic kidney disease (CKD).
Objectives: (1) To determine the incidence of hyperkalemia in dogs with CKD, (2) to determine the proportion of hyperkalemic dogs that required modification of dietary potassium intake, (3) to evaluate the response to dietary modification.
Methods: The hospital database was reviewed retrospectively to identify dogs with CKD and persistent (>5.3 mmol/L on at least 3 occasions) or severe (K ≥ 6.5 mmol/L) hyperkalemia while consuming a therapeutic renal diet. Records of dogs with hyperkalemia that were prescribed a home-prepared, potassium-reduced diet were evaluated further. Response was evaluated by changes in body weight, BCS, and serum potassium concentration.
Results: One hundred and fifty-two dogs were diagnosed with CKD, of which 47% had ≥1 documented episode of hyperkalemia, 25% had ≥3 episodes of hyperkalemia, and 16% had ≥1 episodes of severe hyperkalemia (K > 6.5 mmol/L). Twenty-six dogs (17.2%) with CKD and hyperkalemia were prescribed a potassium-reduced, home-prepared diet. The potassium concentration of all hyperkalemic dogs on therapeutic diets (potassium content, 1.6 ± 0.23 g/1,000 kcal of metabolizable energy [ME]) was 6.5 ± 0.5 mmol/L but decreased significantly to 5.1 ± 0.5 mmol/L in 18 dogs available for follow-up in response to the dietary modification (0.91 ± 0.14 g/1,000 kcal of ME, P < .001). Potassium concentration normalized in all but 1 dog.
Conclusions and Clinical Importance: Hyperkalemia is a potential complication of CKD. In a subset of CKD dogs, hyperkalemia can be associated with commercial renal diets and could restrict use of these diets. Appropriately formulated, potassium-reduced, diets are an effective alternative to correct hyperkalemia.
Association of American Feed Control Officials
angiotensin-converting enzyme inhibitors
body condition score
chronic kidney disease
International Renal Interest Society
maintenance energy requirement
Veterinary Medical Teaching Hospital
Chronic kidney disease (CKD) is a common diagnosis in dogs and occurs in up to 7% of the dogs.1 Hyperkalemia is rarely reported as a clinical feature of dogs with CKD, even though it is a commonly associated complication of acute kidney injury in dogs.2 In recent years, we have observed that hyperkalemia, which might be life threatening, also is associated with CKD because of various mechanisms, 3,4 and thus should be identified and appropriately managed.
The role of dietary intake of potassium in the pathogenesis of hyperkalemia in dogs with CKD has not been fully investigated; nor has reducing the concentration of potassium in the diet been shown to influence serum potassium concentrations in dogs manifesting hyperkalemia. Under some clinical circumstances, the dietary potassium load in therapeutic renal diets could exceed renal excretory capacity resulting in hyperkalemia. Angiotensin-converting enzyme inhibitors (ACEi) are recommended routinely for the management of hypertension and proteinuria, associated with CKD in dogs.5–8 Hyperkalemia is a complication of ACEi therapy in humans with intrinsic kidney disease9,10 and might cause or accentuate a similar and unacceptable risk for hyperkalemia in dogs with CKD that could preclude ACEi use in some dogs.
We hypothesized that feeding a lower amount of dietary potassium daily, as can be achieved by prescribing a diet formulated by a veterinary nutritionist, with food the owner prepares at home, would decrease serum potassium concentrations in dogs that were hyperkalemic despite consuming an appropriate therapeutic renal diet. The objectives of this study were (1) to assess the incidence of hyperkalemia in dogs diagnosed with CKD presenting to the Veterinary Medical Teaching Hospital (VMTH) of the University of California-Davis; (2) to determine among those the percentage of dogs with CKD eating a commercial therapeutic renal diet that prompted prescription of a home-prepared, potassium-reduced diet to manage their hyperkalemia, and (3) to evaluate the clinical and biochemical response in the dogs receiving these dietary prescriptions.
Materials and Methods
Case Selection Criteria
All VMTH records were searched for dogs with a diagnosis of CKD between January 2004 and February 2007. Diagnosis and staging of CKD was performed according to the International Renal Interest Society (IRIS) guidelines and classification scheme.a Dogs manifesting an acute decompensation of their kidney disease <4 weeks before the diet formulation were excluded from the study as were dogs that were dependent on hemodialysis for survival. Records were reviewed to identify dogs with CKD that developed persistent hyperkalemia (serum potassium >5.3 mmol/L on at least 3 separate occasions) or severe hyperkalemia assessed to compromise the animal's health (serum potassium ≥6.5 mmol/L). Medical records of dogs with persistent or severe hyperkalemia while consuming an appropriate commercial therapeutic renal diet that were subsequently prescribed a home-prepared, potassium-reduced diet formulated by a veterinary nutritionist were evaluated retrospectively. Response to the dietary prescription was assessed in these dogs by evaluating retrospectively changes in body weight, body condition score, and biochemical parameters further.b
Medical Records Review
Information extracted from the medical records of the dogs prescribed a home-prepared, potassium-reduced diet included signalment, body weight, body condition score, medical and dietary histories, clinical findings, hematologic and serum biochemistry parameters, and relevant medications. Biochemical parameters including alanine aminotransferase, albumin, alkaline phosphatase, aspartate aminotransferase, anion gap, total calcium, total CO2, chloride, cholesterol, creatinine, globulin, glucose, phosphorus, potassium, sodium, total bilirubin, total protein, blood urea nitrogen, and γ-glutamyltransferase were compared before and 7–14 days after initiation of the home-prepared diet.
All home-prepared, potassium-reduced diets, were formulated by a veterinary nutritionist after consultation with the attending clinician, by 1 of 2 available computer software programs.c,d Each dog received a unique, custom formulation. Ingredients used in each diet formulation varied and were based on potassium content, the dog's ingredient preferences as reported by the owner in the diet history questionnaire, and availability for purchase by the owner. Diets were balanced according to the National Research Council (NRC) and Association of American Feed Control Officials (AAFCO) guidelines for maintenance of adult dogs, except where modified for renal disease. Diets were restricted in protein, phosphorus, and sodium content compared with adult dog maintenance diets and contained increased levels of water-soluble vitamins. Potassium concentration in each formulation was reduced below that provided by the previous commercial therapeutic diet for the given animal.
Each formulation was designed to provide the animal's daily energy intake. Daily energy intake was based on information in the diet history if considered to be reliable, and the dog was in an ideal body condition. Alternatively, if the dog was underweight or if the diet history was considered to be unreliable, the dog's calculated maintenance energy requirement (MER) was used.
Categorical variables were compared by the Fischer exact test. Continuous variables before and after the initiation of home-prepared diet were compared by the paired sample Student's t-test. To evaluate the influence of ACEi on serum potassium, a comparison of potassium concentrations between all dogs that did and did not receive an ACEi was performed by the Student's t-test. Correlation between the dose of ACEi and potassium concentration was evaluated by a Spearman's rank correlation test. Unless otherwise indicated, all data are presented as mean ± SD. P≤ .05 was considered significant. All statistical analyses were performed by NCSS 2000 statistical software.e
One hundred and fifty-two dogs presented to the VMTH for evaluation of CKD between January 2004 and February 2007. Mean serum potassium concentration for all of these dogs at initial evaluation was 4.9 ± 0.8 mmol/L (reference range, 3.6–5.3 mmol/L). Of these, 71 dogs (47%) had at least 1 documented episode of hyperkalemia in which the serum potassium concentration was above the upper limit of the reference range at either the initial evaluation or during the follow-up examination period. Thirty-eight dogs (25%) had ≥ 3 documented episodes of hyperkalemia, and 24 dogs (16%) had at least 1 episode in which the serum potassium concentration was >6.5 mmol/L. Twenty-one dogs (14%) had 1 or more episodes of documented hypokalemia (serum potassium < 3.6 mmol/L). The incidence of hyperkalemia was significantly higher compared with the incidence of hypokalemia (47 versus 14%, respectively, P < .001).
Twenty-six dogs (17.2%) had a nutrition consultation and a prescription formulated by a veterinary nutritionist for a potassium-reduced, home-prepared diet to manage hyperkalemia. The mean age of these dogs was 8.7 ± 4.1 years; 18 were males (12 castrated), and 8 were spayed females. Using the IRIS classification for CKD, 1 dog (4%) was classified as Stage II CKD, 17 dogs (65%) were classified as Stage III, and 8 dogs (31%) were classified as Stage IV CKD.
All 26 dogs for which a potassium-reduced, home-prepared diet was prescribed were consuming 1 of 4 (dry or canned) protein, phosphorus, and sodium modified commercial therapeutic renal diets at the time of diet formulation.f–i Nine dogs (35%) were receiving the commercial diet through an esophagostomy tube at the time the home-prepared diet was prescribed. The mean protein content in the home-prepared diets was 12.9 ± 1.9% on a metabolizable energy (ME) basis, the mean fat content was 26 ± 4.6% on ME basis, and the mean carbohydrate content was 60.9 ± 3.9% on ME basis. The mean potassium content of the 4 commercial therapeutic renal diets was 1.64 ± 0.23 g/1,000 kcal ME compared with a potassium content of 0.91 ± 0.14 g/1,000 kcal ME for the potassium-reduced diets formulated by a veterinary nutritionist.
There were no significant changes in either body weight (21.2 ± 12.2 versus 20.7 ± 12.3 kg, P= .1) or body condition score (4.2 ± 1.6 versus 4.1 ± 1.3 on a 9-point scale, P= .7) for dogs before or during the 7–14-day follow-up period in which the home-prepared diet was fed.
None of the dogs had thrombocytosis either at initial assessment (median platelet count, 198 × 103/μL, range 97–397 × 103/μL; reference range, 150–400 × 103/μL) or during the follow-up period (median platelet count, 176.5 × 103/μL, range, 88–385 × 103/μL) as a potential cause of hyperkalemia. Serum chemistry parameters before and 7–14 days after the initiation of the dietary prescription are presented in Table 1. The serum potassium concentration of dogs consuming a commercial therapeutic renal diet at the time of the nutrition consult was 6.5 ± 0.5 mmol/L (range, 5.7–7.7 mmol/L). Four of the 26 dogs (15%) had a serum potassium concentration between 5.7 and 6.0 mmol/L; 11 dogs (42%) between 6.1 and 6.5 mmol/L; 6 dogs (23%) between 6.6 and 7.0 mmol/L; and in 5 dogs (19%) serum potassium concentration was >7.0 mmol/L. Serum potassium concentration was measured in 18 dogs eating the dietary prescription for 7–14 days. Mean serum potassium concentration decreased significantly from 6.5 ± 0.6 mmol/L to 5.1 ± 0.5 mmol/L (P < .001) with initiation of the dietary prescription and was within the reference range in all but 1 dog whose serum potassium concentration decreased from 6.7 to 6.0 mmol/L but remained outside of the reference range. There were no other statistically significant differences in the serum chemistry results before or after initiation of the potassium-reduced, home-prepared diet.
Table 1. Serum biochemistry parameters before (n = 28) and after (n = 18) consumption of a reduced-potassium, home-prepared diet.
|Alanine aminotransferase (U/L)||60.6 ± 40.9||40.8 ± 25.0||19–67|
|Albumin (g/dL)||2.7 ± 0.4||2.9 ± 0.4||3.0–4.4|
|Alkaline phosphatase (U/L)||107.6 ± 101.5||81.8 ± 42.3||13–190|
|Aspartate aminotransferase (U/L)||31.9 ± 19.2||27.7 ± 5.1||19–42|
|Anion gap (mmol/L)||21.4 ± 4.7||20.3 ± 8.2||10–24|
|Total calcium (mg/dL)||11.7 ± 1.7||11.6 ± 1.8||9.7–11.5|
|CO2 (mmol/L)||20.0 ± 3.9||17.7 ± 4.4||16–26|
|Chloride (mmol/L)||109.7 ± 5.3||108.6 ± 5.8||108–118|
|Cholesterol (mg/dL)||334 ± 120||312 ± 115||135–361|
|Creatinine (mg/dL)||4.2 ± 1.6||4.6 ± 1.8||0.3–1.2|
|Globulin (g/dL)||2.8 ± 0.8||2.6 ± 0.4||1.8–3.9|
|Glucose (mg/dL)||87.3 ± 34.1||98.8 ± 39.3||64–123|
|Phosphorus (mmol/L)||6.3 ± 1.9||5.9 ± 1.2||3.0–6.2|
|Potassium (mmol/L)*||6.5 ± 0.5||5.1 ± 0.5||3.6–5.3|
|Sodium (mmol/L)||145 ± 5||143 ± 5||145–154|
|Total bilirubin (mg/dL)||0.1 ± 0.1||0.1 ± 0.1||0–0.2|
|Total protein (g/dL)||5.6 ± 0.7||5.6 ± 0.4||5.4–7.6|
|Blood urea nitrogen (mg/dL)||61.9 ± 25.2||64.0 ± 30.0||5–21|
|γ-glutamyltransferase (U/L)||4.1 ± 0.9||2.8 ± 0.4||0–6|
Thirteen of the 26 dogs (50%) for which a home-prepared diet was formulated were being treated with an ACEi at a dose of 0.6 ± 0.43 mg/kg/d. Serum potassium concentration was not different between dogs treated with an ACEi as part of their management and dogs that were not receiving an ACEi (6.5 ± 0.62 versus 6.43 ± 0.45 mg/kg/d, P= .6). Nor was there a significant correlation between the dose of ACEi therapy and serum potassium concentration. During the follow-up period there were no significant changes in the mean dose of ACEi before and after the initiation of the home-prepared diet (0.6 ± 0.4 versus 0.5 ± 0.3, respectively, P= 1.00), nor was there a difference in the proportion of dogs receiving an ACEi. During the period the potassium-reduced formulated diet was being fed, the administration of ACEi was discontinued only in 1 dog; the dose was decreased in 2 dogs and increased in 1.
The current study demonstrated that hyperkalemia is a potential complication of dogs with CKD consuming commercial therapeutic diets, but a resolution of the hyperkalemia can be expected in response to potassium-reduced, home-prepared diet formulated by a veterinary nutritionist. In the canine remnant kidney model, oral potassium loading leads to hyperkalemia secondary to the increased intake which is not offset by the decreased kaliuresis in affected dogs compared with normal dogs.11–13 Similarly, dogs with naturally occurring CKD are likely to be at risk for hyperkalemia if exposed to a dietary potassium load that exceeds renal excretory capacity. It has been suggested that residual nephrons in dogs with CKD increase potassium excretion in the distal tubule and maintain potassium homeostasis when fed diets containing up to approximately 0.6% potassium on a dry matter basis until the disease is advanced.14 The average potassium content of the therapeutic renal diets consumed in the current study was 1.6 ± 0.23 g/1,000 kcal ME (range, 1.26–1.9 g/1,000 kcal ME; or 0.54–0.84% on a dry matter basis) and may have contributed to the development of hyperkalemia. The predisposition for hyperkalemia from an excessive dietary load is more likely when a dog consumes enough food to meet its energy requirements. Although one might anticipate hyperkalemia would develop more readily in dogs with advanced kidney disease (IRIS Stage IV), only 8% of dogs in the present study were represented by this stage of CKD. Most dogs (65%) were classified as IRIS Stage III CKD (serum creatinine, 2–5 mg/dL), documenting that hyperkalemia can occur in dogs at a relatively early stage of renal dysfunction. It can be speculated that hyperkalemia was not observed more commonly in dogs with advanced CKD as they frequently failed to consume enough food to meet their energy requirements. Nevertheless, hyperkalemia was still significantly more prevalent than hypokalemia in the entire cohort of dogs evaluated in this study.
Hemodialysis-depended animals were excluded from this study because of the severity of their renal dysfunction, but, in our experience, dogs with either acute kidney injury or CKD managed with hemodialysis invariably develop hyperkalemia, which is often life threatening when managed with commercial therapeutic renal diets.j It is now recognized that different stages of CKD require different levels of medical intervention.14 In the same manner, diets appropriate for dogs with early stages of CKD might not be appropriate or well suited for dogs with more advanced stages of disease. Progressive and proportional reduction of dietary potassium should be considered as part of the therapeutic strategy for dogs with advanced kidney disease to obviate development of hyperkalemia.
As GFR declines in dogs with progressive CKD, clinical signs including anorexia or vomiting become more pronounced, food consumption decreases, and weight loss occurs. At this point many clinicians and owners feed human foods and over-the-counter maintenance diets, which are not formulated for the management of CKD to prevent further weight loss and starvation in an attempt to maintain an appropriate energy intake. However, feeding foods and diets not intended for animals with CKD will likely accelerate disease progression, worsen the clinical signs of uremia, and promote biochemical abnormalities such as hyperphosphatemia.15 The use of an indwelling feeding tube can facilitate administration of the animal's daily energy and nutrient requirements with diets designed for the management of CKD. A potential adverse consequence of tube feeding to achieve caloric adequacy is the delivery of some nutrients in the diet, like potassium, in excess of the animal's intake if it did not eat to caloric adequacy and in excess of the kidney's excretory capacity. In this study, 9/26 dogs (35%) were managed with the aid of a feeding tube. This more aggressive approach might have been responsible, in part, for the high prevalence of dogs requiring a potassium-reduced, home-prepared diet. Conversely, it likely facilitated the management of these dogs and extended their life expectancy. It should be noted, however, that 65% of the dogs in this study became hyperkalemic while consuming commercial veterinary therapeutic renal diets without the aid of a feeding tube, suggesting that commercial veterinary therapeutic renal diets can contribute to hyperkalemia in some dogs with CKD.
In this study only serum potassium changed after the initiation of a potassium-reduced, home-prepared diet formulated by a veterinary nutritionist. The absence of a difference in serum total CO2 concentrations reduced the likelihood that changes in acid base status influenced transcellular shifts of potassium to promote the decrease in serum potassium concentrations. Similarly, an influence of serum glucose and, by inference, insulin concentrations are unlikely to be responsible for the differences in potassium concentrations between time points. Mineralocorticoids were not measured, but the normal serum electrolyte concentrations, exclusive of potassium, argue against a role for alterations in mineralocorticoids. Catecholamines cannot be excluded as the reason for the difference in serum potassium concentrations, but it is unlikely it could have a major influence on serum potassium during the brief follow-up period. The dose of ACEi was only changed in 4 dogs during the follow-up period and, therefore, is unlikely to account for the observed decrease in serum potassium for the entire treated cohort during the follow-up period when the potassium-reduced diet was fed. For all of the aforementioned reasons, the decrease in serum potassium concentrations can be most likely attributed to the change in dietary potassium intake.
ACEi are commonly used in the management of CKD in dogs; thus it is crucial to maintain serum potassium concentrations within the reference range to allow the safe use of ACEi.16,17 In our study, 50% of dogs were being treated with a relatively low dose of an ACEi (mean 0.6 ± 0.43 mg/kg/d) at the time of diet formulation. Because of the nature of this retrospective study, it was difficult to determine whether this dose was adequate, or if the hyperkalemia prevented the use of higher doses. In the present study, there was no apparent influence of either the use or dose of ACEi therapy on mean serum potassium concentration in the dogs for which a home-prepared diet was formulated, but the lack of influence may have been caused by the relatively low doses reported.
Dietary potassium may not be the sole factor contributing to hyperkalemia in our study population, and it is possible the combination of advanced kidney disease, aggressive dietary management, dietary potassium intake, and concurrent medical therapy cumulatively predisposed these dogs to the risk of hyperkalemia. It also is difficult to assess the relative contribution of each of these potential factors. Nevertheless, all dogs but 1 dog showed normalization of serum potassium concentrations within 2 weeks of initiating a potassium-reduced formulated diet, suggesting dietary potassium was a major contributor to the hyperkalemia. Because of the nature of this retrospective study, data regarding food consumption during the follow-up period compared with the preceding period was not available, and it cannot be ruled out entirely that some dogs consumed less food during the follow-up period, resulting in a decrease in serum potassium concentrations secondary to a reduction in dietary potassium. However, body weight and body condition scores remained unchanged in the dogs following the institution of the potassium-reduced diets, suggesting energy intake was not reduced.
Both therapeutic renal diets and ACEi have been shown to have benefits in dogs with CKD, but likewise can become contraindicated in dogs that develop hyperkalemia on these therapies.15,16 These results verify that a potassium-reduced, home-prepared diet appropriately formulated for dogs with CKD by a trained veterinary nutritionist can resolve the hyperkalemia that may otherwise preclude ongoing management of the renal disease with an ACEi. The reported AAFCO minimum recommendation for potassium in commercial food is 1.7 g/1,000 kcal of ME, and the recommendation for potassium intake in dogs with CKD has been suggested to be 0.8–1.2 g/1,000 kcal of ME.18,19 Evidence supporting a dietary recommendation for potassium for dogs with CKD is lacking to date, but the suggestion of 0.8–1.2 g/1,000 kcal of ME is comparable to the content of the potassium-reduced, home-prepared diets that resolved the hyperkalemia in this study. The mean potassium content of the renal therapeutic diets consumed by dogs before diet formulation was 1.64 ± 0.23 g/1,000 kcal of ME (range, 1.26–1.9 g/1,000 kcal ME), which is nearly twice the concentration demonstrated to maintain normal serum potassium concentrations in the current study. The prevalence and severity of hyperkalemia demonstrated in dogs with CKD in the current study and their response to dietary potassium reduction suggest the concentration of potassium found in current veterinary therapeutic renal diets may be inappropriate for some animals and would justify a dietary prescription by a trained veterinary clinical nutritionist.
Nutritionists often recommend a veterinary therapeutic diet to address a medical concern, if appropriate commercial therapeutic diets are available. There are, however, a number of reasons to prescribe an individually formulated, home-prepared diet for some animals. The major indication for feeding an animal a formulated diet prepared by the owner is a medical condition that has special or unique nutritional concerns not appropriately or specifically addressed by a commercial therapeutic product. Home-prepared diets are often the only option for animals with multiple medical conditions. By selecting a therapeutic diet to treat 1 condition, the practitioner may be feeding in a manner that is appropriate for 1 condition and contraindicated for others.
Prescribing a home-prepared diet for an animal is not without risk. It is important to remember that home-prepared diets typically have not undergone animal feeding trials or even laboratory analysis to confirm that they support the life-stage for which they were designed. Owners often substitute ingredients without first consulting a veterinary nutritionist and thereby alter the diet in such a way that might potentiate or harm the dog's underlying condition. It is equally important to obtain the prescription for a therapeutic home-prepared diet from a reputable source, and ideally the diet should be formulated by a trained veterinary nutritionist who understands both the nutritional and physiological aspects of the animal's medical condition. Clinicians should be extremely cautious of recipes obtained from the Internet or publications designed for use by the general public, which have not been formulated for a specific animal or by a veterinary nutritionist. Many of these sources lack detailed information regarding supplements and cooking methods that require the owner to make decisions that often change the diet extensively and in many cases in a manner that is not appropriate or in the best interests of the dog. A number of these resources are also out of date with respect to current nutrition knowledge and research.
aIRIS staging as accepted by the European Society for Veterinary Nephrology and Urology
bHitachi 717, Roche Laboratories, Indianapolis, IN
cSmall Animal Nutritionist, N Squared Computing & Durango Software, Silverton, OR
dBalance IT Professional, DVM Consulting, Prof Corp, Davis, CA
eNCSS 6.0.22 software, NCSS, Kaysville, UT
fEukanuba Veterinary Diet Canine/Advance Stage Renal, The Iams Company, Dayton, OH
gHill's Prescription Diet Canine k/d. Hill's Pet Nutrition Inc, Topeka, KS
hPurina Veterinary Diets Canine NF, Nestlé Purina PetCare Company, St Louis, MO
iRoyal Canin Veterinary Diet Canine Renal LP, Royal Canin Inc, St Charles, MO
jPantaleo V, Francey T, Cowgill LD. Analysis of hyperkalemia in dogs on chronic hemodialysis. ACVIM Forum, 2005, Baltimore, MD (abstract)