Abstract
Practical relevance:
Diabetic ketoacidosis (DKA) is a not uncommon emergency in both newly diagnosed and poorly regulated diabetic cats. When there is a heightened metabolic rate and energy requirement due to concurrent illness, an increase in the release of glucose counter-regulatory hormones causes insulin receptor resistance, lipolysis, free fatty acid release and ketogenesis. This necessitates not only treatment to eliminate the ketosis and control blood glucose, but also investigation of concurrent illnesses.
Clinical challenges:
A number of metabolic derangements can occur with DKA, requiring a comprehensive diagnostic evaluation, elimination of ketones, careful correction of glucose, electrolyte and acid base abnormalities, and close monitoring.
Audience:
Any veterinarian that cares for cats in urgent and emergency situations should understand the pathophysiology of DKA in order to address an individual’s clinical signs and metabolic derangements.
Evidence base:
This review draws evidence from the peer-reviewed literature as well as the author’s personal clinical experience.
DKA – care plan essentials
Cats may live with diabetes mellitus (DM) without showing overt signs of illness for a period of weeks to months. However, any increase in metabolic rate and energy requirement due to concurrent illness produces an increase in the release of glucose counter-regulatory hormones, resulting in ketogenesis. Significant ketonemia overwhelms the plasma buffering system, and produces a metabolic acidosis that can alter normal cellular metabolism. The clinical signs are related to ketoacidosis, hyperosmolarity and concurrent illness.
Feline diabetic ketoacidosis represents a potentially fatal condition that is diagnosed with a combination of findings including hyperglycemia, glucosuria, ketonemia or ketonuria, and a metabolic acidosis. The mainstay of therapy is immediate recognition and treatment of life-threatening problems, careful volume replacement, and insulin administration for glycemic control and elimination of ketonemia. different insulin protocols exist for the acute care of the cat with DKA. Electrolyte imbalances are common sequelae, and need to be recognized and corrected. Likewise, concurrent conditions need to be recognized and treated. Prior to initiation of intensive care, the client must be appraised of the long-term management requirements for the cat with diabetes mellitus, including twice-daily insulin injections and increased veterinary visits.
Pathophysiology
Glucose is a necessary fuel that drives the production of energy, in the form of adenosine triphosphate (ATP), by the mitochondria within the cells. In all organs, other than the brain, insulin is needed to move glucose from the blood into the cells where it is used to make ATP or is converted and stored as glycogen (in the liver and muscles) or triglyceride (in adipocytes). Insulin also moves electrolytes such as potassium from the plasma into the cells and limits the oxidation of fat.
DM is a syndrome caused by insulin insufficiency, insulin receptor dysfunction, insulin receptor downregulation, or some combination that results in lack of transport of glucose into the cell, hyperglycemia and cellular starvation. Pancreatic islet cell destruction and concurrent disorders that affect insulin receptor sensitivity (eg, pancreatitis, chronic inflammatory diseases, obesity, endocrinopathies) are conditions that are most commonly associated with feline DM.1,2 Hyperglycemia that exceeds the renal threshold for glucose reabsorption results in glucosuria, increased urine osmolality and urine production, as well as increased plasma osmolarity and thirst. Cellular starvation stimulates an increase in appetite, mobilization of free fatty acids (FFAs), an increase in plasma amino acids and weight loss. oxidation of FFAs leads to hepatic production of ketone bodies (acetoacetate, beta-hydroxybutyrate and acetone), which are used as an alternative energy source in the peripheral tissues and liver.
Acquired Dm can be tolerated for a period of time. Since many cats free-feed, their waste may not be cleaned daily, and/or they have indoor–outdoor access, clients may fail to immediately recognize the classic signs of DM (increased food intake, drinking and urination). As long as there is adequate, albeit reduced, circulating insulin, hyperglycemia does not lead to acute illness. However, when there is an increased metabolic rate and energy requirement due to concurrent illness, the release of glucose counter-regulatory hormones (adrenaline [epinephrine], norepinephrine, cortisol, glucagon and growth hormone) is stimulated. These ‘stress’ hormones increase insulin receptor resistance, lipolysis and FFA release, and promote ketogenesis. Significant ketonemia related to DM overwhelms the plasma buffering system, and results in a metabolic acidosis (ie, DKA) that can alter normal cellular metabolism. Similar to glucose molecules, ketones are osmotically active. Increased filtration of ketones by the kidney exacerbates osmotic diuresis, electrolyte imbalances and water loss, manifesting in hypovolemia and interstitial dehydration. Untreated DKA causes severe illness and death.
Clinical signs and diagnosis
Cats suffering from DKA typically present with problems of an urgent or emergency nature. Client complaints are usually nonspecific and may include lethargy, anorexia and vomiting. Weakness and gait abnormalities may also be observed. 3 A complete history may uncover more classic signs of hyperglycemia including recent increased drinking, urination and appetite, with weight loss. Physical examination findings can include clinical signs of poor perfusion, altered mentation, dehydration and poor body condition. In some cases, a sweet smell to the breath can alert the examiner to the presence of exhaled acetone.
Laboratory confirmation of DKA is based on the presence of hyperglycemia, glucosuria, ketonemia or ketonuria, and a metabolic acidosis. Serum fructosamine can be evaluated in cases with marginal glucose levels.
Triage and fluid resuscitation
The primary survey performed during patient triage will identify any immediate life-threatening problems. The breathing effort is characterized, and the lungs are carefully auscultated for evidence of increased or absent breath sounds. Perfusion is assessed, and body temperature obtained. It is common for poor perfusion in the cat to manifest as pale mucous membranes, prolonged capillary refill time, bradycardia, hypotension and hypothermia. 4 When altered mentation exists, the cat must be carefully handled to minimize stress and increases in intracranial pressure. Oxygen is supplied while immediate placement of a peripheral intravenous (iV) catheter and collection of blood and urine samples for analysis is performed. Intramuscular (iM) or IV injection of 0.4–0.6 mg/kg butorphanol may reduce patient anxiety associated with restraint.
Resuscitation from hypotension always involves the use of an isotonic crystalloid solution. Some clinicians prefer to use 0.9% sodium chloride (308 mOsm/l) to provide the least degree of transcellular osmolar shift when severe hyperglycemia is present, and avoid the use of lactated Ringer’s solution to reduce competition for hepatic conversion of lactate and ketones. However, 0.9% sodium chloride is an acidic fluid and may not correct an acidemia as effectively as a balanced buffered isotonic solution such as Plasma-Lyte A pH 7.4 (Baxter) or Normosol-R pH 7.4 (Hospira) (~295 mOsm/l), which contain acetate and gluconate buffers that are metabolized by skeletal muscle. It is not uncommon for hyponatremia to be present, and some clinicians might also choose to infuse 0.9% sodium chloride to correct this. Hyponatremia, however, can be a spurious finding when hypertriglyceridemia exists, and/or it can be due to shift of water intravascularly in response to increased plasma osmolality caused by hyperglycemia (see box on page 1172).
Resuscitation of the hypovolemic cat differs from that of the dog. 4 Large volume resuscitation, as typically used in the dog, may result in volume overload in the cat. A more measured approach using intermittent bolus infusions of body temperature 10–15 ml/kg isotonic crystalloid solution, with or without 3–5 ml/kg hydroxyethyl starch (HES), can be administered over 10–15 mins while monitoring the arterial blood pressure. Repeated boluses are administered until the systolic arterial blood pressure is at least 40–60 mmHg. At this stage, active external rewarming is initiated by placing the cat in a warming tent, on a circulating warm water/air blanket, and/or running the fluid infusion line through a warm fluid bath or fluid warmer (Figure 1). Fluid administration is continued at calculated maintenance and rehydration rates until the rectal temperature reaches 98°F (36.7°C). If hypotension still exists after rewarming, crystalloid and HES boluses are continued. If hypotension persists in the normothermic cat following the infusion of 60 ml/kg buffered isotonic crystalloid and 20–40 ml/kg HES, causes of non-responsive shock must be investigated, and vasopressors may be indicated.
Figure 1.

Critically ill cats with hypothermia must be carefully warmed following initiation of fluid resuscitation. This patient is covered with a blanket to prevent heat loss, and active warming is being provided by placement of the fluid line through a warming device
Analysis of packed cell volume, total protein, electrolytes, venous blood gas, glucose, osmolality and lactate data is performed. Samples are saved for additional evaluation (complete blood count, serum biochemical profile, urine analysis and culture, and thyroid profile) after the cat is stabilized. Heparinized plasma can be tested for ketones using urine reagent strips (Figure 2). 5 Ketostix reagent strips use a colorimetric method that measures a nitroprusside reaction for detecting acetoacetate in blood or urine; blood measurements are more precise than urine measurements. 6 This method does not, however, measure beta-hydroxybutyrate (which can be present before detectable levels of acetoacetic acid), making it insensitive for monitoring the severity of ketoacidosis. Thus monitoring of beta-hydroxybutyrate is preferred, and persistent positive results are considered in conjunction with clinical signs when transitioning from ICU to home care. Ketone testing using the Precision Xtra or Optium systems (Abbott) will measure beta-hydroxybutyrate.
Figure 2.

Reagent strips are used to test the urine for glucose and ketones. A serum blood sample can also be tested for ketones using the reagent strip. Courtesy of Kate Hopper
ICU management
Once perfusion is restored, rehydration therapy is instituted (see box on page 1170). At the point when the cat is better hydrated, a multilumen central venous catheter can be placed in the jugular or median saphenous vein to facilitate repeated blood collection and central venous pressure monitoring (Figure 3). Since frequent blood collection in the cat can lead to anemia, 8 some clinicians prefer to collect blood from the marginal ear veins for periodic glucose checks with a glucometer (Figure 4). A portable glucometer (AlphaTRAK 2; Zoetis) is specifically calibrated for use in cats and dogs (Figure 5). It requires only a 0.3 μl sample, and blood from a marginal ear vein prick using a 25 G needle is sufficient.
Figure 3.

Placement of a central venous catheter allows repeated blood sampling for glucose monitoring without having to restrain or repeatedly prick the patient with a needle. The same catheter can be used for fluid and drug administration. Courtesy of Elisa Mazzaferro
Figure 4.

Marginal ear vein sampling for glucose monitoring. Courtesy of Zoetis
Figure5.

The AlphaTRAK 2 glucometer, specifically calibrated for cats and dogs, requires only 0.3 μl of blood, and is suitable for home monitoring. The monitor’s reading range is 20–750 mg/dl. A ‘HI’ reading indicates blood glucose is >750 mg/dl; a ‘LOW’ reading indicates it is <20 mg/dl. Courtesy of Mandy Nonnemacher
Table 1.
Physical examination findings used to estimate interstitial dehydration
| Estimated % dehydration | Physical examination findings |
|---|---|
| 4–6 | ● Tacky mucous membranes |
| 6–8 | ● Loose skin turgor
● Dry mucous membranes |
| 8–10 | ● Loose skin turgor
● Dry mucous membranes ● Globes retracted within orbits |
| 10–12 | ● Persistent skin tent due to complete loss of skin elasticity
● Dry mucous membranes ● Retracted globes ● Dull corneas |
| >12 | ● Persistent skin tent
● Dry mucous membranes ● Retracted globes ● Dull corneas ● Signs of perfusion deficits (eg, pale mucous membranes, prolonged capillary refill time, poor pulse quality, hypothermia, bradycardia) |
Insulin therapy and protocols
Insulin infusion is started after perfusion has been restored. There is a misconception that insulin should not be initiated until dehydration has been corrected, to mitigate the impact of dilution and increased glomerular filtration, which risk rapidly decreasing the effective osmolality and also potentially causing cerebral edema. 9 However, a retrospective study in 60 dogs and cats with DKA or diabetic ketosis showed that a delay to insulin therapy of >6 h significantly delayed resolution of ketonemia (by 19 h, on average), and no difference in complications was observed. 10
There are several methods for insulin administration in the cat with DKA. Whatever method is prescribed, the ultimate goal is to eliminate ketosis and correct acidosis without causing hypoglycemia.
Intermittent administration of regular insulin Administration of regular insulin can be successfully employed at an initial dosage of 0.1 U/kg IM followed by 0.05 U/kg IM administered hourly until the blood glucose level is <250 mg/dl. 11
Glargine insulin IM glargine (which has a short half-life) administered with or without subcutaneous (SC) glargine (which has a long half-life) has also been shown to treat feline DKA effectively, and may be considered in cases with limited finances, when continuous insulin infusion and in-hospital monitoring may not be affordable. In a retrospective study reviewing treatment of DKA in 15 cats, 12 an initial dose of 1–2 U per cat was administered IM with 1–3 U SC, followed by 1–2 U IM as needed (q2h or less frequently) and 1–2 U SC q12h until regulated. Based on the results of their study, the authors recommend the use of 1–2 U SC glargine q12h, with 0.5–1 U IM up to q4h, to achieve a blood glucose concentration of 180–252 mg/dl. Inadequate perfusion and absorption from depot sites may make this method of administration less predictable, and it also requires more frequent blood glucose checks (q1–2h) compared with a continuous infusion method.
Continuous infusion of regular insulin Use of a constant rate infusion (CRI) of regular insulin permits a more regulated decline in serum glucose levels and plasma osmolarity because adjustments in infusion rates can be made. Douglass Macintire first introduced the concept of insulin continuous infusion for veterinary patients in a hyperglycemic crisis in 1993, 9 and it has become the standard of care in many veterinary ICUs. Her recommendation for a feline patient is to place 1.1 U/kg of regular insulin in a 250 ml bag of 0.9% saline, drain out 50 ml of the solution to allow for insulin adsorption by the plastic tubing, and administer the mixture at a rate of 10 ml/h.
A higher dose of insulin infusion (up to 2.2 U/kg q24h) may be more effective at reducing hyperglycemia in the cat and is associated with a better outcome than a lower dose. 13 There does not appear to be a relationship between osmolality at presentation, administered insulin dose and mentation changes. 14
A pilot study involving 29 cats with DKA compared the concurrent use of regular insulin IM (1 U up to q6h) and SC glargine (0.25 U/kg q12h) with a continuous infusion of regular insulin (1 U/kg q24h) in cats with DKA. 15 The former regimen produced more rapid resolution of ketosis, hyperglycemia and acidemia, and an associated reduction in the length of hospital stay. This author has used concurrent regular insulin continuous infusion using the above rate of 1 U/kg q24h and SC glargine (1 U q12h) in cats with success.
Patient monitoring and additional treatment
Following the initiation of volume replacement and insulin therapy, close monitoring is needed to recognize alterations in fluid balance, maintain perfusion and hydration, prevent hypoglycemia, and identify and treat persisting or developing acid–base and electrolyte abnormalities (see box on page 1172). Treatment can become complicated when there is a need for supplemental electrolytes and glucose. Placing fluids with additives in a burette drip chamber permits adjustment of additive concentrations without sacrificing an entire bag of fluids.
Patient monitoring includes evaluating blood glucose q2–4 h, and electrolytes, venous blood gases, fluid ins and outs, and body weight q12–24h (using the same scales at each weigh-in to avoid discrepancies). The ideal blood glucose level is between 150 mg/dl and 250 mg/dl. Regular insulin continuous infusion therapy may be adjusted according to glucose levels (Table 2). Traditionally, the recommendation has been to supplement dextrose when the blood sugar is <250 mg/dl,9,11 but this can become costly when frequent changes in supplementation (including electrolytes) necessitate additional fluid bags, IV administration sets and burette drip chamber sets. It also may increase the time it takes to determine when the patient may be ready to transition from a CRI to twice-daily insulin injections. This author’s experience is that, with adequate monitoring, adjustments can be made to the insulin infusion, and the addition of dextrose may not be necessary until the blood glucose drops below 100 mg/dl.
Table 2.
Adjustment of regular insulin CRI based on blood glucose levels
| Blood glucose | Treatment changes | |
|---|---|---|
| (mg/dl) | (mmol/l) | |
| >400 | >22 | If glucose is not declining after two or three rechecks, ↑ insulin CRI by 25% |
| 250-400 | 13.9-22 | Continue as initially planned |
| 200-250 | 11.1-13.9 | ↓ Insulin CRI by 25% |
| 150-200 | 8.3-11.1 | ↓ Insulin CRI by 25% |
| 100-150 | 5.5-8.3 | ↓ Insulin CRI by 25% |
| <100 | <5.5 | Stop insulin infusion, start 2.5% dextrose IV. If clinical signs exist, slowly bolus 0.1-0.25 g/kg 50% dextrose (0.5-1 ml/kg) and repeat blood glucose after 30 mins |
CRI = constant rate infusion; IV = intravenously
Table 3.
Supplementation rates for hypokalemia
| Patient’s serum potassium concentration (mEq/l) | Suggested supplementation dose (mEq KCI/I) | Suggested fluid rate (ml/kg/h) |
|---|---|---|
| 3.1–3.5 | 28 | 18 |
| 2.5–3.0 | 40 | 12 |
| <2.5 | 60 | 8 |
| <2.0 | 80 | 6 |
Based on the patient’s cardiovascular and volume status, additional monitoring may include periodic evaluation of arterial blood pressure, body temperature, hydration level and body weight.
Broad spectrum intravenous antibiotic therapy (eg, ampicillin-sulbactam 20 mg/kg IV q8h) may be indicated when a left shift is identified, a fever is present, severe gastrointestinal signs exist, or there are overt signs of a bacterial infection. Thoracic radiographs, abdominal ultrasound and urine culture are used as screening tools for identifying pulmonary, intra-abdominal or genitourinary infections, respectively.
Nasogastric tube feeding, with or without partial parenteral nutrition, is necessary to preserve enterocyte function, reverse protein catabolism and promote hepatic function until the patient is eating voluntarily. Assisted feeding also limits the development of food aversion, and the cephalic and gastric phases of digestion that can promote vomiting in cats with nausea. Nasogastric tube placement permits evaluation of gastric emptying function with periodic aspiration, as well as immediate administration of liquid nutrition. CliniCare (Zoetis) can be infused as a continuous infusion starting with a 50% solution at 0.5 ml/kg/h. Over a 48 h period, this is increased to a 100% solution at ~2 ml/kg/h, provided that gastric residual volumes are minimal. Emeraid Intensive Care (Emeraid) can also be bolused at 2–6 h intervals. Prokinetic medication such as metoclopramide and/or cisapride can promote gastric emptying.
FreAmine (B Braun) is an intravenous 3% amino acid solution that can be used as a daily maintenance fluid and provides partial parenteral nutrition when administered at a maintenance fluid rate once the patient has started insulin therapy. ProcalAmine (B Braun) is another intravenous amino acid (3–4%) solution that contains glycerin as a carbohydrate source. Both solutions contain potassium and can be administered via a peripheral catheter. Since they are given as a continuous infusion, additives such as antiemetics and vitamin B can be administered with these fluids.
Additional evaluation
Ketoacidosis, by itself, has not been associated with a poor outcome in the diabetic cat; 21 however, coexisting conditions (see box below) may affect the prognosis and outcome, and should be identified and addressed. Serum biochemistry, thyroid function testing, a complete blood count, and urine analysis and culture are indicated. Although not predictive of mortality, elevations in serum creatinine, blood urea nitrogen, magnesium and total bilirubin are associated with a worse outcome in the cat with DKA. 13 Thoracic and abdominal radiographs as well as abdominal ultrasound are necessary for uncovering evidence supporting liver disease, pancreatitis, kidney abnormalities, infection and neoplasia. When liver enzymes are elevated and/or biliary changes exist, liver aspirate/biopsy collection for cytopathological and culture analysis may be indicated. A common concurrent disease finding in cats is hepatic lipidosis, which may require additional therapeutic measures. 22 When chronic gastrointestinal (GI) signs occur, endoscopic biopsy of GI mucosa may uncover infiltrative diseases.
Preparing for hospital discharge
Once the cat is hydrated and voluntarily eating or tolerating tube feeding, and the ketosis is cleared, long-term insulin therapy can be initiated. 21 The insulin CRI is discontinued at least 4 h prior to starting long-acting insulin injections, unless the cat is already receiving insulin glargine. The blood glucose is evaluated at the time insulin injections are to be given, as determined by the owner’s schedule. If the serum glucose is <150 mg/dl, insulin is not administered. At the next scheduled time for insulin therapy, the glucose level is checked. On the basis it is >250 mg/dl, 0.5–1 U glargine or protamine zinc insulin q12h SC is administered. This dose is recommended in the previously diagnosed diabetic, as well as newly diagnosed cases, since insulin requirements may have altered.
A glucose curve is evaluated over the following 12 h and adjustments are made to insulin injections as necessary. The goal is to have a glucose nadir no lower than 150 mg/dl in the hospital, rather than try to determine the ideal dose for long-term management. When prescribing insulin at the time of discharge, consideration should be given to the fact that the environmental conditions at home will be less stressful than in the hospital, and therefore a lower dose of insulin may be required than was given in hospital.
The patient is discharged with instructions for feeding and for home urine glucose monitoring (Figure 6), with a glucose curve evaluation scheduled for 1 week later. 23 The client is instructed to call a veterinarian if the urine glucose is repeatedly negative or >2000 mg/dl. A glucose curve and/or serum fructosamine estimation are needed to determine any future adjustments in insulin therapy. ‘Spot check’ glucose tests are only useful if hypoglycemia is suspected (Figure 7), since a single glucose value that is normal or increased cannot differentiate adequate therapy from too much or too little insulin. For example, a cat receiving too much insulin can become hypoglycemic, then experience a rebound hyperglycemia related to stress hormone release (ie, Somogyi effect); if the blood glucose is measured during this rebound, the resulting hyperglycemia can be misinterpreted as inadequate insulin administration. Increasing insulin in such cases can cause life-threatening hypoglycemia.
Figure 6.

Reagent strips specific for measuring urine glucose and ketone can be used by owners for home monitoring. Courtesy of Steve Epstein
Figure 7.

A single glucometer reading showing an elevated blood glucose level. ‘Spot check’ glucose tests are only useful if hypoglycemia is suspected (see text). Courtesy of Megan Tremelling
Key Points
Essential treatment for the cat with DKA includes judicious fluid replacement, insulin administration and correction of electrolyte imbalances. Ketonemia will not resolve without insulin administration.
Insulin therapy for patient stabilization can be customized as long as blood glucose is closely monitored.
Treatment of DKA in itself is relatively straightforward. What must be included in the care plan for the cat with DKA is investigation and treatment of concurrent illnesses that cause increased counter-regulatory hormone release.
Clients must be prepared to treat their diabetic cat with twice-daily insulin and additional veterinary visits prior to committing to treatment of the cat with DKA.
Footnotes
The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding: The author received no financial support for the research, authorship and/or publication of this article.
References
- 1. Feldman EC, Nelson RW. Feline diabetes mellitus. in: Feldman EC, Nelson RW. (eds). Canine and feline endocrinology and reproduction. 3rd ed. Philadelphia: Elsevier Science, 2004, pp 539–579. [Google Scholar]
- 2. Rand JS. Pathogenesis of feline diabetes. Vet Clin North Am Small Anim Pract 2013; 43: 221–231. [DOI] [PubMed] [Google Scholar]
- 3. Kerl ME. Diabetic ketoacidosis: treatment recommendations. Compend Contin Educ Pract Vet 2001; 23: 330–339. [Google Scholar]
- 4. Rudloff E, Kirby R. Feline circulatory shock. In: Little S. (ed). August’s consultations in feline internal medicine. Vol 7. St Louis, MO: Elsevier, 2016, pp 843–858. [Google Scholar]
- 5. Brady MA, Dennis JS, Wagner-Mann C. Evaluating the use of plasma hematocrit samples to detect ketones utilizing urine dipstick colorimetric methodology in diabetic dogs and cats. J Vet Emerg Crit Care 2003; 13: 1–6. [Google Scholar]
- 6. Zeugswetter F, Pagitz M. Ketone measurements using dipstick methodology in cats with diabetes mellitus. J Small Anim Pract 2009; 50: 4–8. [DOI] [PubMed] [Google Scholar]
- 7. Rudloff E. Assessment of hydration. In: Silverstein D, Hopper K. (eds). Textbook of small animal critical care. Philadelphia: WB Saunders, 2014, pp 307–331. [Google Scholar]
- 8. Lynch AM, Respess M, Boll AE, et al. Hospital-acquired anemia in critically ill dogs and cats: a multi-institutional study. J Vet Intern Med 2016; 30: 141–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Macintire DK. Treatment of diabetic ketoacidosis in dogs by continuous low-dose intravenous infusion of insulin. J Am Vet Med Assoc 1993; 202: 1266–1272. [PubMed] [Google Scholar]
- 10. DiFazio J, Fletcher DJ. Retrospective comparison of early-versus late-insulin therapy regarding effect on resolution of diabetic ketosis and ketoacidosis in dogs and cats: 60 cases (2003-2013). J Vet Emerg Crit Care 2016; 26: 108–115. [DOI] [PubMed] [Google Scholar]
- 11. Feldman EC, Nelson RW. Diabetic ketoacidosis. In: Feldman EC, Nelson RW. (eds). Canine and feline endocrinology and reproduction. 3rd ed. Philadelphia: Elsevier Science, 2004, pp 580–615. [Google Scholar]
- 12. Marshall RD, Rand JS, Gunew MN, et al. Intramuscular glargine with or without concurrent subcutaneous administration for treatment of feline diabetic ketoacidosis. J Vet Emerg Crit Care 2013; 23: 286–290. [DOI] [PubMed] [Google Scholar]
- 13. Cooper RL, Drobatz KJ, Lennon EM, et al. Retrospective evaluation of risk factors and outcome predictors in cats with diabetic ketoacidosis (1997-2007): 93 cases. J Vet Emerg Crit Care 2015; 25: 263–272. [DOI] [PubMed] [Google Scholar]
- 14. Claus MA, Silverstein DC, Shofer FS, et al. Comparison of regular insulin infusion doses in critically ill diabetic cats: 29 cases (1999-2007). J Vet Emerg Crit Care 2010; 20: 509–517. [DOI] [PubMed] [Google Scholar]
- 15. Gallagher BR, Mahony OM, Rozanski EA. A pilot study comparing a protocol using intermittent administration of glargine and regular insulin to a continuous rate infusion of regular insulin in cats with naturally occurring diabetic ketoacidosis. J Vet Emerg Crit Care 2014; 25: 1–6. [DOI] [PubMed] [Google Scholar]
- 16. Riordan LL, Schaer M. Potassium disorders. In: Silverstein DC, Hopper K. (eds). Small animal critical care medicine. 2nd ed. St Louis, MO: Elsevier, 2015, pp 269–273. [Google Scholar]
- 17. Bateman S. Disorders of magnesium: magnesium deficit and excess. In: DiBartola SP. (ed). Fluid therapy in small animal practice. 3rd ed. Philadelphia: WB Saunders, 2006, pp 210–226. [Google Scholar]
- 18. Barton L, Kirby R. Electrolytes. In: Kirby R, Linklater A. (eds). Monitoring and intervention for the critically ill small animal: the rule of 20. Ames: John Wiley & Sons, 2017, pp 73–94. [Google Scholar]
- 19. Barton L. Magnesium and phosphate disorders. In: Silverstein DC, Hopper K. (eds). Small animal critical care medicine. 2nd ed. St Louis, MO: Elsevier, 2015, pp 281–288. [Google Scholar]
- 20. Statz G. Feline diabetic ketoacidosis. In: Little S. (ed). August’s consultations in feline internal medicine. Vol 7. St Louis, MO: Elsevier, 2016, pp 836–842. [Google Scholar]
- 21. Callegari C, Mercuriali E, Hafner M, et al. Survival time and prognostic factors in cats with newly diagnosed diabetes mellitus: 114 cases (2000-2009). J Am Vet Med Assoc 2013; 243: 91–95. [DOI] [PubMed] [Google Scholar]
- 22. Armstrong PJ, Blanchard G. Hepatic lipidosis in cats. Vet Clin North Am Small Anim Pract 2009; 39: 599–616. [DOI] [PubMed] [Google Scholar]
- 23. Sparkes AH, Cannon M, Church M, et al. ISFM consensus guidelines on the practical management of diabetes mellitus in cats. J Feline Med Surg 2015; 17; 235–250. [DOI] [PMC free article] [PubMed] [Google Scholar]






