Insulin therapy in dogs and cats (Proceedings)

Insulin has become increasingly important in the treatment of dogs and cats with diabetes mellitus. Almost all dogs are classified with insulin-dependent or type 1 diabetes mellitus. Although most cats are non insulin-dependent or type 2 diabetics at the time of diagnosis, early intervention with exogenous insulin is important component to achieving remission and preserving the function of insulin-secreting beta cells. For this reason almost all of our diabetic dogs and cats will be treated with insulin so it is important to understand which insulin preparations are available and their utility in our canine and feline patients.

History

Insulin was first discovered in 1921 by Frederick Banting and colleagues. The original preparations were bovine and porcine pancreatic extracts. These formulations were short-acting, relatively dilute and impure. As a result, frequent administration and large volumes of insulin were required. Due to impurities, local reactions were common. Because beef and pork insulin differ from human insulin by 3 and 1 amino acids, respectively, many people developed insulin antibodies. Over time more concentrated beef and pork insulins were produced and purity improved. In the 1930's through 1950's it was discovered that the addition of protamine and zinc would cause crystallization of insulin and slow absorption. In the 1980's recombinant DNA technology led to the development of human recombinant insulins. The production of human recombinant insulin lead to a dramatic decrease in the use of animal-origin insulins in man. The 1990's lead to the discovery of synthetic insulin analogues in an attempt to achieve better glycemic control in people.

Ultra short and short – Acting insulins (lispro, aspart, regular)

Insulin analogues were developed to better mimic the basal-bolus requirements of man as well as improve consistency. The premise is that a basal amount of insulin is continuously produced by the pancreas as a result of glucose produced by the liver (glycogenolysis, gluconeogenesis). There is also a larger, more rapid increase (bolus) in insulin after meals as a result of post-prandial hyperglycemia.

Human recombinant regular insulin was the traditional 'bolus' insulin used separately or in mixed preparations with an intermediate-acting insulin. Using regular insulin, peak insulin levels occurred at 60 minutes with a duration of up to 6 hours. This is because regular insulin exists as a hexamer (6 insulin molecules) that dissociates into dimers and monomers once injected subcutaneously. Monomers are the active form and required for absorption from subcutaneous sites into the bloodstream. When given intravenously regular insulin rapidly dissociates to monomers so there is a more rapid onset and shorter duration. Because of the pharmacokinetics of subcutaneously administered human recombinant regular insulin in man, it is administered 30 to 45 minutes prior to eating. This was inconvenient so ultra short-acting analogues were developed that could be administered at the time of or 15 minutes prior to eating. Insulin analogues manipulate the amino acids sequences to alter absorption and dissociation. The ultra-short analogues achieved a more rapid onset, higher peaks and shorter duration. Insulin lispro (Humalog® - Eli Lilly & Co.) was created by switching proline-lysine to lysine-proline at the 28th and 29th positions on the beta chain. Insulin aspart (Novolog® - Novo Nordisk Laboratories) was created by the substitution of aspartic acid for proline at the 28th position on the beta chain. These substitutions prevent self-aggregation of the insulin molecules that can delay absorption. These ultra short-acting analogues do not appear to be of any clinically significant benefit over human recombinant regular insulin in dogs and cats.

Intermediate - Acting insulins (NPH, Lente)

In 1946 Krayenbuhl and Rosenberg of Novo-Hagedorn Laboratories, following the work of Scott and Fisher (see PZI below), discovered that the addition of equal amounts of a protein derived from fish sperm, protamine, and regular insulin with a small amount of zinc prolonged the duration of insulin. This led to the development of Neutral Protamine Hagedorn (NPH). Neutral Protamine Hagedorn insulin is also referred to as isophane insulin due to its symmetrical structure and equal composition of regular insulin and protamine. This is the only human recombinant insulin that can be mixed with regular insulin because the regular insulin is not bound by the NPH so in mixtures (25/75, 50/50) there is a short duration of effect derived from the regular and a longer effect from the NPH insulin, respectively. Neutral Protamine Hagedorn must be given to cats and dogs twice daily. The peak is often too rapid and duration too short in both species, even with twice daily administration. This is particularly problematic in cats so this insulin is not routinely used in this species.

In 1952 Hallas-Moller developed lente insulin when they discovered that the addition of zinc and an acetate buffer crystallized insulin and prolonged its duration. Lente insulin is actually 30% of short-acting semilente (not commercially available) and 70% long-acting ultralente. Initial preparations were beef/pork products because the pork insulin tends to be amorphous whereas the beef insulin crystallizes to a greater extent. The greater amount of zinc in lente insulin causes micro crystallization of any added regular insulin so they are not mixed. Currently a pork lente insulin (Vetsulin®, Caninsulin® - Intervet/Schering Plough) is available in the US as a U-40 preparation. Pork insulin is considered ideal for the dog because porcine and canine insulin are identical. Porcine lente has also been used successfully in cats for many years.

Long – Acting insulins

In 1936 the work of Hagedorn, Scott and Fisher led to the development of protamine zinc insulin (PZI). Hagedorn discovered that the addition of various proteins, including protamine, could prolong insulin activity. Scott and Fisher also added the zinc and PZI was born. Eli Lilly & Co. manufactured PZI until 1991 at which time this insulin was unavailable commercially in the U.S. for several years. Commercial production resumed of a beef/pork formulation through IDEXX (PZI VET®) until 2008 when there was no longer an FDA approved source of beef extract available. There is still beef/pork PZI insulin available through distributors but to my knowledge commercial production in the U.S. has ceased once again. There have been studies recently evaluating a human recombinant PZI in cats that are promising so this formulation may become available shortly. Because PZI has historically been primarily beef in origin, there were concerns that dogs would develop antibodies thus it has not been routinely recommended in dogs. Despite being referred to as a long-acting insulin, the majority of cats can not be regulated on once daily dosing. This is also a U-40 insulin so special syringes are required.

More recently, long-acting insulin analogues have been developed to more closely mimic the basal insulin secretion in man. Historically, the human recombinant intermediate insulins had to be given to people twice daily, which was inconvenient. Human recombinant insulins also showed patient-to-patient, interpatient and site variability in effect. For this reason the long-acting analogues, such as glargine (Lantus® - Aventis Pharmaceuticals) and detemir (Levemir® - Novo Nordisk), were developed utilizing recombinant DNA technology.

Insulin glargine was created by substituting a glycine amino acid for an asparagine at the 21st position of the alpha chain. Two arginines were also added to the beta chain. This results in an insoluble insulin molecule at a neutral pH. It is solubilized at an acidic pH and once administered into the neutral sub cutis becomes insoluble to slow absorption. Glargine is relatively peakless, exhibits little patient-to-patient variability and can be given once daily in man. Because it is formulated in an acidic pH it can not be diluted or mixed. Dogs have shown a less predictable response to glargine than cats so it is not commonly used in this species. Glargine has become increasingly popular in cats with diabetes due to increased rates of diabetic remission compared to lente and PZI insulin in newly diagnosed diabetic cats. Glargine has a longer duration than lente or PZI and can be used once daily in cats. Although many cats can be controlled with once daily administration, twice daily administration is believed to provide superior glycemic control and improve remission rates, particularly in newly diagnosed diabetic cats. The initial dose recommendations in cats are 0.5 U/kg BID if glucose is >/= 360 mg/dL and 0.25 U/kg if glucose is < 360 mg/dL. Because hypoglycemia is fairly common when instituting therapy, glucose curves (pre insulin then every 4 hours) are recommended for the first 72 hours when instituting therapy. The recommended dose in dogs is 0.25 U/kg twice daily. Glargine can be stored in the refrigerator for 6 months.

Insulin detemir is another long-acting analogue created by the addition of a fatty acid moiety to position 29 of the beta chain. This insulin is absorbed rapidly but then the fatty acid moiety binds circulating albumin and slows activity. There is limited experience with this analogue in dogs and cats. Dosing recommendations in the dog and cat are similar to glargine. Although starting doses are the same as glargine, Dr. Jacque Rand reports that final doses of detemir in cats may be approximately 30% less than glargine.

Complications of insulin therapy

Persistence of clinical signs is common with insulin therapy. This may be a result of problems with the insulin itself, insulin under dosage, insulin over dosage or insulin resistance. Insulin may be ineffective if it is stored, handled or administered incorrectly. Unused insulin is typically refrigerated. Once opened, it is often stored at room temperature when used in man. This led to the recommendation that insulin be replaced every 30 days due to potential loss of potency as well as contamination. It is recommended that insulin for animals is stored in the refrigerator once opened. This is thought to prolong potency and prevent contamination. The exact duration of maximum potency for most insulin in dogs and cats as well as people is unknown. Most insulin contains crystals that if disturbed can lead to alterations in absorption and potency. For this reason insulin suspensions (NPH, PZI, lente) should not be shaken. It is typically recommended to administer insulin over the lateral thorax and flanks alternating sites. Skin over the neck and back may be thickened and insulin deposited intradermally affecting absorption. Animals can occasionally develop changes in the skin and dermis at the site of repeated injection and this might interfere with absorption. Dilute insulins can have variable potency so dilution should be avoided if possible. If diluting insulin, the diluent recommended by the manufacturer should be used.

Insulin under dosage should be suspected in a dog given < 1.0 U/kg or a cat given < 0.5 U/kg with persistence of clinical signs (weight loss, increased water consumption, increased urination, persistently elevated urine glucose, urinary ketones). Clinical signs with high serum fructosamine and/or inadequate response on a blood glucose curve supports under dosage. Under dosage is more common in animals on once daily insulin as a result of short duration of effect.

Hypoglycemia is fairly common in diabetic dogs and cats that undergo insulin therapy but not all animals are symptomatic. With profound hypoglycemia (glucose < 65 mg/dL) diabetogenic hormones are released that result in increased glucose. Glucose may increase to over 400 mg/dl and remain increased for many hours. This protective mechanism is blunted in some individuals resulting in hypoglycemia. Hypoglycemia is more common in newly diagnosed diabetics, cats with transient diabetes, animals given large increases in dose of insulin, patients on twice daily long-acting insulin, during increased activity and with decreased caloric consumption. Clinical signs of hypoglycemia include inappetence, lethargy, weakness, seizures and coma. Normal fructosamine and repeatedly negative urine glucose are findings that should also raise suspicions of hypoglycemia.

Insulin resistance has classically been defined as poor glycemic control at >1.5 U/kg/dose in the dog and cat with most animals controlled on < 1 U/kg/dose. Transient insulin resistance can occur due to hyperglycemia as part of insulin over dosage via the Somogyi phenomenon but may also occur as the result of insulin antagonistic drugs, diseases or conditions. Bacterial infections (in particular the oral cavity and urinary tract), pancreatitis, hyperadrenocorticism, drugs (e.g. glucocorticoids, progestins), hyperthyroidism (cat), stress (cat) and obesity are relatively common causes of resistance. Chronic inflammation, acromegaly (cat), hypothyroidism (dog), renal disease, liver disease, cardiac disease, glucagonoma, pheochromocytoma, exocrine pancreatic insufficiency, hyperlipidemia and other forms of neoplasia have also been reported to cause insulin resistance. With insulin resistance glucose curves may show little to no response to insulin or a response may be noted but excessive amounts of insulin required.

Insulin is a foreign protein that can result in anti insulin antibodies. Antibodies are thought to result from differences in conformational epitopes of insulin molecules. The formation of antibodies results in erratic changes in blood glucose, insulin resistance and poor glycemic control. In dogs and cats, antibodies to human recombinant insulin are uncommon and occur in less than 5%. Dogs do tend to develop antibodies to beef insulins, with an incidence as high as 45% with beef/pork formulations.

Adjustments in insulin therapy

It may take dogs and cats several days to adapt to the introduction of insulin. For this reason hyperglycemia should not be concerning in the newly diagnosed healthy diabetic and increases in maintenance insulin are not typically recommended earlier than one week after instituting therapy. Patients may be monitored after starting insulin therapy for hypoglycemia and appropriate reductions in insulin dose made. As stated previously, hypoglycemia is fairy common with initiation of glargine insulin in cats but clinical signs of hypoglycemia are rare. In general if hypoglycemia develops in dogs and cats, the dose of insulin is reduced by 25 - 50%. In some cases (e.g. Somogyi with excessive insulin doses) it may be advisable to revert to a recommended initial dose. Remember in cats diabetic remission is possible so in those on small amounts of insulin (</= 0.5 unit per dose) that develop hypoglycemia; discontinuation of insulin may be indicated.

Traditionally diabetics have been monitored with clinical signs (weight, water consumption, urine production, glucosuria, ketonuria), glucose curves and serum fructosamine levels. Clinical signs continue to be important in evaluating a diabetic patient. A well-regulated animal should no longer be polyuric, polydipsic and achieve ideal body weight. Ketones are occasionally seen in the urine of a healthy diabetic dog but should not be a persistent finding. Persistently high levels of glucose in the urine (3 or 4+) is also suggestive of poor control.

In cats initially placed on glargine or detemir, glucose is sometimes monitored on initiation and insulin doses adjusted for hypoglycemia and prolonged hyperglycemia (> 300 mg/dl for 72 hours) but in most instances curves are not performed earlier than 7 days after initiation of therapy. Curves are then performed every 7 to 14 days until a stable insulin dose has been determined. With in-hospital glucose curves, it is difficult to get animals to mimic their normal activity level and caloric consumption. In-hospital glucose curves are also problematic in cats due to stress hyperglycemia. The impact of these factors can be minimized by having owners perform glucose curves at home. With intermediate-acting, more potent insulins (NPH, lente), curves most commonly include a pre-administration glucose followed by glucose measurements every 2 hours until the next insulin dose is due. For the less potent, 'peakless' insulins (glargine, detemir) a pre-administration glucose followed by measurements every 3 to 4 hours until the next insulin dose is due is usually adequate.

It is difficult to make general recommendations regarding insulin adjustments in dogs and cats. One of the most important factors when making adjustments in insulin dose is the presence of clinical signs. It is also important to remember that human glucometers are designed to measure venous blood glucose in people and are not calibrated for use in dogs and cats. Previously established guidelines were made utilizing human glucometers which can underestimate glucose in dogs and cats. The following are recommendations utilizing glucometers calibrated for dogs and cats. If utilizing human glucometers, these recommendations should be adjusted with a decrease of 30 mg/dl.

Overall goals of therapy

1. Diabetic remission in cats

2. Control of clinical signs

3. Pre-insulin glucose < 250 mg/dl (dog), < 300 mg/dl (cat)

4. Nadir 100 to 150 mg/dl

Pre-insulin glucoses higher than 250 mg/dl in the dog and 300 mg/dl in the cat in the presence of a high nadir > 150 to 170 mg/dl should prompt an increase in insulin dose. Increases in the dog depend on the current dose and degree of glucose elevation but do not usually exceed 20 to 25%. In the cat increases are dependent on the current dose, degree of glucose elevation and insulin type but are usually in 0.25 to 1 unit increments. If the pre-insulin glucose is < 200 mg/dl in the dog or cat, the nadir and clinical signs should be evaluated. In combination with a nadir < 80 mg/dl, a reduction in insulin dose is indicated, particularly if clinical signs of hypoglycemia are present. Reductions are dependent on the nadir and presence of clinical signs but are typically 50%. When weaning cats off insulin it is recommended they be gradually decreased to 0.5 units per dose and then insulin discontinued. If the pre-insulin glucose is < 200 mg/dl and the nadir is 100 to 150 mg/dl then no adjustment is necessary. If the pre-insulin glucose is < 200 mg/dl and the nadir is > 150 mg/dl then the insulin dose might be too high, diabetic remission (cat) may have occurred, insulin antagonism may have subsided or there may be prolonged duration of effect. Regarding duration, if giving twice daily insulin and the nadir occurs within 3 hours or the duration is < 8 hours a longer-acting formulation should be considered. If the duration is greater than 12 hours (nadir > 8 hours) the options are once daily administration or twice daily administration at a lower dose. Helpful references for adjustments in insulin therapy can be found in Feldman EC, Nelson RW Canine and Feline Endocrinology and Reproduction as well as the companion animal health website at the University of Queensland's http://www.uq.edu.au/ccah/index.html?page=41544

Fructosamine is a measure of glycated proteins in the serum and is a reflection of glucose levels over 1 to 2 weeks time. Fructosamine can also help eliminate concerns of stress hyperglycemia in cats but as discussed above there are many factors that affect fructosamine so using them as a sole guide for insulin therapy is not recommended. Signs of hyperglycemia in the face of an elevated fructosamine should prompt a glucose curve to assess glycemic control.

References

Feldman EC, Nelson RW. Canine diabetes mellitus in Canine and Feline Endocrinology and Reproduction. Feldman EC, Nelson RW eds. 2004: pp 486-538.

Feldman EC, Nelson RW. Feline diabetes mellitus in Canine and Feline Endocrinology and Reproduction. Feldman EC, Nelson RW eds. 2004: pp 539-579.

Leahy JL. Intensive insulin therapy in type 1 diabetes mellitus in Insulin Therapy Leahy JL, Cefalu WT eds. 2006; pp 87 – 112.

Marshall RD, Rand JS, Morton JM. Treatment of newly diagnosed diabetic cats with glargine insulin improves glycemic control and results in higher probability of remission than protamine zinc and lente insulins. J Feline Med Surg. 2009 Aug;11(8):683-91.

Michiels L, Reusch CH, Boari A, et al. Treatment of 46 cats with porcine lente insulin—a prospective, multicentre study. J Feline Med Surg 2008 Oct;10(5):439-51.

Nelson RW, Henley K, Cole C, et al. Field safety and efficacy of protamine zinc recombinant human insulin for treatment of diabetes mellitus in cats. J Vet Int Med 2009 Jul-Aug;23(4):787-93.

Norsworthy GD, Lynn R, Cole C. Preliminary study of protamine zinc recombinant insulin for the treatment of diabetes mellitus in cats. Vet Therap 2009 Spr-Summer; 1-2(10):24-28.

Plum A, Agerso H, Andersen L. Pharmacokinetics of the rapid-acting insulin analog, insulin aspart, in rats, dogs, and pigs, and pharmacodynamics of insulin aspart in pigs. Drug Metab Dispos. 2000 Feb; 28(2): 155-60.

Ritzel RA, Butler PC. Physiology of glucose homeostasis and insulin secretion in Insulin Therapy Leahy JL, Cefalu WT eds. 2006; pp 61 – 71.

Weaver KE, Rozanski EA, Mahoney OM, et al. Use of glargine and lente insulins in cats with diabetes mellitus. J Vet Int Med 2006 Mar-Apr;20(2):234-8.

Wittlin SD, Woehrle HJ, Gerich JE. Insulin pharmacokinetics in Insulin Therapy Leahy JL, Cefalu WT eds. 2006; pp 73 – 86.