A range of insulin choices and management strategies are available for our veterinary patients. Recently, a new type of insulin, glargine, has become available, and many owners are now able to monitor blood glucose concentrations at home. This article reviews these options and provides guidelines for starting therapy in stable, nonketotic dogs and cats.
Persistent fasting hyperglycemia is the hallmark of diabetes mellitus. In dogs, you can confidently diagnose diabetes mellitus if a patient has a fasting serum glucose concentration > 200 mg/dl. As the threshold for glucose resorption from the glomerular filtrate is 180 to 220 mg/dl, concurrent glycosuria will be present.1
Cats can become transiently hyperglycemic (serum glucose concentration > 300 mg/dl) when stressed, so a single elevated serum glucose concentration, even with concurrent glycosuria, is not enough to diagnose diabetes mellitus. Supportive clinical signs such as weight loss, polyuria, and polydipsia must be documented. In equivocal cases, measuring the serum fructosamine concentration (which reflects serum glucose levels for the previous three weeks) may be necessary. Alternatively, diagnostic litter additives (e.g. Glucotest—Nestlé Purina) can be used to document persistent glycosuria in the home environment. Unless a cat is ketoacidotic, insulin therapy should not be started without clear evidence of persistent hyperglycemia.
Once diabetes mellitus has been confirmed, certain tests are indicated to direct appropriate therapy and to check for concurrent disease. Many patients are presented to their veterinarian after many weeks of hyperglycemia, and it is often an acute complication, such as pancreatitis or a urinary tract infection, that prompts the owner to seek help. Failure to recognize concurrent problems will affect the patient's response to therapy and may ultimately affect the outcome.
It is essential to determine whether a new diabetic is ketotic. A routine urine dipstick test will readily identify acetone and acetoacetate; the urine must be mixed with hydrogen peroxide to facilitate detection of beta-hydroxybutyrate, which is often the most prevalent ketone body. In early ketosis, renal and respiratory compensation prevent substantial acidosis, but these systems soon become overwhelmed, and dehydration and compromised tissue perfusion then exacerbate the situation. Once diabetic ketoacidosis is established, aggressive in-hospital care becomes necessary, but this is beyond the scope of this article.
Because glucose is a nutrient source for microorganisms, many diabetic patients present for evaluation of bacterial urinary tract infections. It is important to remember that severe polyuria may make it difficult to identify bacteria and white blood cells on a urine sediment examination. A urine culture with antimicrobial sensitivity testing is always indicated, but start appropriate antibiotics immediately if infection is suspected.
Serum chemistry profile
Concurrent problems, such as renal failure, pancreatitis, and hyperadrenocorticism, can profoundly affect a patient's response to insulin. A serum chemistry profile is necessary to identify other diseases and to provide a baseline for future comparison. Closely evaluate markers for dehydration, such as azotemia or hyperproteinemia, and provide fluid therapy if necessary. Serum cholesterol and triglyceride concentrations may be elevated as a result of the diabetes but can also suggest concurrent thyroid or adrenal dysfunction. The serum sodium concentration is often decreased in response to glucose-related hyperosmolality; this hyponatremia does not need to be specifically addressed in an adequately hydrated patient. Total body stores of potassium and phosphate are often low, despite normal serum concentrations. In dehydrated or anorectic patients, these parameters need close monitoring, and supplementation is necessary to prevent precipitous decreases as insulin is introduced.
Complete blood count
A diabetic patient that seems otherwise healthy is expected to have normal complete blood count results. Anemia (which may be masked by dehydration) or leukocytosis should prompt further investigation.
Screen all cats over 6 years of age for hyperthyroidism by measuring a total thyroxine (T4) concentration. If the concentration is at the upper end of the normal range, early hyperthyroidism may be more readily identified by measuring the free T4 concentration by equilibrium dialysis.3 Dogs with untreated diabetes may have low total T4 concentrations because of euthyroid sick syndrome. So evaluate a free T4 concentration and thyroid-stimulating hormone concentration before starting thyroid supplementation.
Choosing an insulin
The cost of insulin varies. The veterinary-licensed products are generally more expensive than human ones and require particular syringes (designed for U-40 insulin) that are not available at human pharmacies.
The following is intended as a general description of insulin products commonly used in cats and dogs. It does not cover every formulation presently available, and readers are urged to carefully research any unfamiliar type before starting therapy. Doses are provided here as a guide and must always be tailored to a patient's specific requirements.
Regular insulin. Regular insulin is a human recombinant type available in a U-100 concentration (e.g. Humulin R—Eli Lilly, Novolin R—Novo Nordisk). Regular insulin is highly potent and is generally reserved for in-hospital use in ketoacidotic animals. It is the only insulin that can be administered intravenously. Careful monitoring is necessary to prevent a precipitous drop in serum glucose concentrations, and it is not an appropriate choice for at-home therapy.
Vetsulin. This Lente insulin from Intervet, known as Caninsulin outside the United States, is of porcine origin and comes in a U-40 concentration. It is an aqueous zinc suspension of amorphous and crystalline insulin, which produces two peaks of activity—one soon after administration and the second several hours later.4
Vetsulin is FDA-approved for use in dogs only. It can be used in cats, but insulin resistance secondary to anti-insulin antibodies is a theoretical concern in this species. Some dogs may be adequately controlled with once-daily administration, but most dogs require twice-daily dosing.5 I recommend a conservative starting dose of 0.5 U/kg, to minimize the risk of hypoglycemia. Doses should be adjusted based on serial glucose measurements: increase by 10% to 15% in a persistently hyperglycemic patient (glucose > 250 mg/dl), and decrease by 25% if hypoglycemia (glucose < 70 mg/dl) is noted.
NPH insulin. This form of insulin is a human recombinant type and is available in a U-100 concentration (e.g. Humulin N—Eli Lilly, Novolin N—Novo Nordisk). NPH is an intermediate-duration insulin, with modest potency. It can be used in both cats and dogs and is an economical option in large dogs (when compared with Vetsulin). Twice-daily dosing is usually necessary, and a starting dose of 0.5 U/kg is generally appropriate.6,7 As with Vetsulin, dosages should be adjusted based on serial glucose measurements: increase by 10% to 15% in a persistently hyperglycemic patient (glucose > 250 mg/dl), and decrease by at least 25% if hypoglycemia (glucose < 70 mg/dl) is noted.
Protamine zinc insulin. Protamine zinc insulin (PZI) is 90% bovine origin and 10% porcine origin and is available in a U-40 concentration that is FDA-approved for use in cats (PZI Vet—Idexx). It is a long-acting insulin and is an appropriate choice for diabetic cats. Some cats achieve acceptable glycemic control with once-daily dosing, but twice-daily dosing is more common. The recommended starting dose is 1 U/cat (0.22 to 0.6 U/kg). The dose can be increased in increments of 0.5 U if necessary. PZI is not suitable in dogs because its onset and duration of effect are unpredictable and there is a risk of anti-insulin antibody formation.8
A surprisingly large number of newly diagnosed diabetic cats receiving glargine go into remission, most likely because of the reversal of glucose toxicity,9 a condition in which persistent hyperglycemia inhibits the release of insulin from the pancreatic beta cells. If the glucose concentrations are quickly normalized, the beta cells regain function, and the diabetes may resolve.
This insulin cannot be diluted, as its effect is pH-dependent. Dosing accuracy is improved if 0.3-ml 100-U syringes are used. It is available in 10-ml vials and 3-ml cartridges; for cost reasons, some clients prefer to purchase the cartridges and have the insulin decanted by a pharmacist into an appropriate container.
It may take a few weeks to find the appropriate dose for a patient, and even well-controlled diabetics usually need periodic adjustments. For dogs, I recommend a recheck after the first seven days of insulin therapy and then every six to eight weeks. Any time the insulin dose is changed, reevaluate the patient within two weeks.
A veterinary-specific glucose monitor (AlphaTRAK—Abbott Laboratories) is now available and provides rapid glucose measurement with a single drop of blood. A 22-ga needle or a lancet device is used to puncture the skin, generally on the pinna or lip, and the capillary action strip draws the sample into the machine. Many clients can easily learn to do this at home and are able to check their pets' blood glucose concentrations without the inconvenience and expense of a visit to the veterinarian's office. Cats particularly benefit from at-home monitoring, as stress hyperglycemia is avoided. Even if clients are reluctant to check glucose concentrations on a regular basis, the ability to measure a glucose concentration if the pet is unwell or acting strangely can be a lifesaver.
I do not encourage owners to change insulin doses based solely on at-home glucose measurements. Rather, the results should be added to other information such as weight, thirst, hunger, urine output, and energy levels before dose adjustments are made.
If owners are unwilling or unable to measure blood glucose concentrations at home, they can purchase urine dipsticks for glucose and ketone detection. If ketones are frequently noted, the patient should be rechecked by the veterinarian. Most dogs with well-regulated diabetes have mild glycosuria most of the day; any patient with persistently high urine glucose concentrations or negative urine glucose concentrations is probably receiving an inappropriate insulin dose and should be examined. Do not instruct owners to increase the insulin dose if substantial glycosuria is noted, as many patients that are receiving too much insulin have periods of rebound hyperglycemia with subsequent spillage of glucose into the urine. Instead, owners should regard this as an indication for a visit to the hospital and a full evaluation.
Serial glucose assays provide the most detailed evaluation of a patient's response to insulin. Food and insulin should be given on the usual schedule, and blood should be drawn for glucose measurement every two hours. If the glucose drops precipitously, more frequent monitoring may be necessary to identify the nadir. For patients receiving once-daily insulin, the glucose curve should be continued for 24 hours; for patients receiving twice-daily therapy, it is acceptable to measure glucose concentrations over 10 to 12 hours. Checking the glucose concentration every four hours is generally sufficient in patients receiving glargine since glargine has a slow onset of action (Figure 3).
Ideally, blood glucose concentrations should stay within the normal range, but this rarely occurs. As veterinary patients do not suffer from the retinal, renal, and vascular complications seen in people, blood glucose concentrations do not need to be as rigidly controlled. However, even mild persistent hyperglycemia can cause cataracts (in dogs) and peripheral neuropathy (in cats). Hypoglycemia should be avoided, and prolonged or extreme hyperglycemia is undesirable. As a general rule, if the lowest blood glucose concentration recorded in the clinic is < 70 mg/dl, the insulin dose should be reduced by 25%. If the glucose concentration is more than 250 mg/dl on two or more readings, a 10% increase in insulin is appropriate. It is important to remember that acute and severe hypoglycemia can trigger rebound hyperglycemia (the Somogyi effect), so never use a single high glucose measurement as an indication to increase the insulin dose.
Watching the client give a dose to the patient can be informative. Some people shake the bottle vigorously, causing damage to the molecules, or inadvertently draw air into the syringe. When handling issues have been addressed, it is necessary to look for other health problems that may affect the body's response to insulin. It is also worthwhile to make sure the patient isn't receiving any other medications (including topical corticosteroids) that may antagonize the effect of insulin.
The impact of dietary formulation on canine diabetes appears to be modest. A recent study evaluating the effect of dietary fiber on postprandial glycemia did not demonstrate a significant effect on blood glucose concentrations,10 although other studies have indicated improved fructosamine concentrations in diabetic dogs fed high-fiber diets.11,12 A palatable balanced diet is essential, but therapeutic diets may not provide substantial clinical advantages over commercial dog food. It is more important for feeding schedules and food type to remain consistent. I recommend offering food immediately before insulin administration to ensure the patient is interested in eating. In addition, for patients receiving Vetsulin, the first peak of activity provides an excellent buffer against postprandial hyperglycemia.
Diabetic cats are quite different because dietary formulation may facilitate reversal of the diabetes.13,14 Low-carbohydrate and high-protein diets are recommended (e.g. Purina Veterinary Diets DM—Nestlé Purina, Prescription Diet m/d—Hill's), although other medical problems such as kidney disease may limit dietary manipulations. As cats can be finicky, finding a well-tolerated and palatable balanced diet is the most important consideration. Canned food tends to be lower in carbohydrates than dry food, but dry kitten chow is an acceptable option if a cat declines one of the therapeutic diabetic diets. It isn't necessary to meal-feed cats receiving PZI or glargine, as both are well-suited to their natural tendency to eat small amounts frequently. Foods high in simple sugars, such as semi-moist diets and treats, should be avoided. Obesity is a well-recognized cause of insulin resistance in cats, and gradual weight loss may improve glycemic control and encourage endogenous insulin production.
Oral hypoglycemic therapy
Dogs with diabetes mellitus are always insulin-dependent, so the oral hypoglycemic drugs are never appropriate. Although some cats respond to glipizide (thought to enhance endogenous insulin production), most are not adequately controlled, and the prolonged hyperglycemia may contribute to glucose toxicity and reduce the chance of eventual diabetic remission.15 I would only prescribe glipizide if a client refused to administer insulin, and I would combine it with appropriate dietary modification and weight loss (if necessary). Acarbose (an enzyme inhibitor that impairs intestinal glucose absorption) may facilitate glycemic control in cats fed a low-carbohydrate diet.16 However, the effect seems modest, at best, and it cannot be used in place of insulin.
Most dogs and cats with diabetes are successfully managed and continue to be a source of joy and comfort to their families. Effective client communication is essential, as pet owners need to feel supported by the veterinary team, particularly at the time of diagnosis. It is important to stress the benefits of regular rechecks and to explain that dose adjustments are a routine part of diabetic care. Encouraging owners to check glucose concentrations at home can be empowering, and many Web sites offer guidance and support (e.g. petdiabetes.org, caninediabetes.org, felinediabetes.com). Reinforce to owners that diabetes mellitus is a treatable disease, and most patients enjoy an excellent quality of life.
Audrey K. Cook, BVM&S, MRCVS, DACVIM (internal medicine), ECVIM-CA
Department of Small Animal Clinical Sciences
College of Veterinary Medicine and Biomedical Sciences
Texas A&M University
College Station, TX 77843
1. Nelson RW. Diabetes mellitus. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 6th ed. St. Louis, Mo: Elsevier Saunders, 2005;1563-1591.
2. Rand JS, Marshall RD. Diabetes mellitus in cats. Vet Clin North Am Small Anim Pract 2005;35(1):211-224.
3. Peterson ME, Melian C, Nichols R. Measurement of serum concentrations of free thyroxine, total thyroxine, and total triiodothyronine in cats with hyperthyroidism and cats with nonthyroidal disease. J Am Vet Med Assoc 2001;218(4):529-536.
4. Graham P, Nash AS, McKellar QA. Pharmacokinetics of porcine insulin zinc suspension in diabetic dogs. J Small Anim Pract 1997;38(10):434-438.
5. Horn B, Mitten RW. Evaluation of an insulin zinc suspension for control of naturally occurring diabetes mellitus in dogs. Aust Vet J 2000;78(12):831-834.
6. Melendez L, Lorenz M. Canine diabetes mellitus, in Proceedings. Western Vet Conf 2002.
7. Bruyette D. Diabetes mellitus: treatment options, in Proceedings. World Small Anim Vet Assoc World Congress 2001.
8. Davison LJ, Ristic JM, Herrtage ME, et al. Anti-insulin antibodies in dogs with naturally occurring diabetes mellitus. Vet Immunol Immunopathol 2003;91(1):53-60.
9. Rand J. Editorial: glargine, a new long-acting insulin analog for diabetic cats. J Vet Intern Med 2006;20(2):219-220.
10. Ferguson DC, Hoenig M, Klaser DA, et al. Glucose tolerance and lipid profiles in dogs fed different fiber diets. Vet Ther 2001;2(2):160-169.
11. Kimmel SE, Michel KE, Hess RS, et al. Effects of insoluble and soluble dietary fiber on glycemic control in dogs with naturally occurring insulin-dependent diabetes mellitus. J Am Vet Med Assoc 2000;216(7):1076-1081.
12. Graham PA, Maskell E, Rawlings JM, et al. Influence of a high fibre diet on glycaemic control and quality of life in dogs with diabetes mellitus. J Small Anim Pract 2002;43(2):67-73.
13. Bennett N, Greco DS, Peterson ME, et al. Comparison of a low carbohydrate-low fiber diet and a moderate carbohydrate-high fiber diet in the management of feline diabetes mellitus. J Feline Med Surg 2006;8(2):73-84.
14. Frank G, Anderson W, Ballam J, et al. Use of a high-protein diet in the management of feline diabetes mellitus. Vet Ther 2001;2(3):238-246.
15. Feldman EC, Nelson RW, Feldman MS. Intensive 50-week evaluation of glipizide administration in 50 cats with previously untreated diabetes mellitus. J Am Vet Med Assoc 1997;210(6):772-777.
16. Mazzaferro EM, Greco DS, Turner AS, et al. Treatment of feline diabetes mellitus using an alpha-glucosidase inhibitor and a low-carbohydrate diet. J Feline Med Surg 2003;5(3):183-189.