Difficult-to-Regulate Diabetic Cats
ACVIM 2008
Anthony C.G. Abrams-Ogg, DVM, DVSc, DACVIM (SAIM)
Guelph, ON, Canada


The primary goal in treating diabetic cats is to regulate blood glucose (BG) levels. What constitutes poor regulation depends on what is considered acceptable regulation. For this discussion, difficult regulation in a cat is defined as failure to consistently control hyperglycemia, using a normal dose of an oral hypoglycemic agent (OHA) or insulin, for a sufficient part of the day to minimize PU/PD, normalize appetite, promote weight gain and activity, and lower fructosamine ± glycosylated hemoglobin levels.

Difficult regulation occurs in three main forms: 1) Consistently bad days with persistent hyperglycemia; 2) Inconsistent regulation, where the cat has good and bad days (the extreme is "brittle" diabetes with recurrent fluctuations between hyperglycemia/ ketoacidosis and hypoglycemia); and 3) insulin resistance (IR), where unusually high doses of insulin are required to achieve regulation. While clinical signs and fructosamine levels are frequently sufficient to confirm good regulation or to document poor regulation, BG curves may be required to identify the cause of poor regulation. Difficulty in obtaining reliable BG curves is one reason why diabetic cats have been difficult to regulate. The use of continuous glucose monitoring systems may help in this regard.1-2

Causes of difficult regulation include: 1) problems with dietary therapy; 2) problems with OHA therapy; 3) problems with insulin therapy; and 4) disorders causing IR.More than one of these may be occurring in an individual case. The diagnostic approach to the poorly-regulated cat should be systematic and the potential causes critically reviewed for each case, but the specific approach will vary with the individual case depending upon the extent of initial evaluation for concurrent diseases, clinical signs, current therapy, and owner desires. In some cases knowing that concurrent/underlying diseases have been ruled-out by exhaustive screening will facilitate patience with adjustments in therapy. In other cases, where there are no obvious signs of a concurrent disorder, an owner may wish to try adjustments in therapy prior to extensive diagnostic testing. Ideally only one change in management should be made at a time.

Problems with Dietary Therapy

The diet may be improper. There is mounting evidence that reducing carbohydrate and increasing protein content of the diet will improve regulation (and increase the likelihood of resolution of diabetes),3-5 and early dietary change is recommended for cats with persistent hyperglycemia without another obvious cause. It is not known if dietary change will always improve inconsistent regulation and reduce insulin requirements when there is therapeutic insulin resistance, but it is a rational manoeuvre.

Feeding schedule is important in humans in dogs, primarily because of post-prandial hyperglycemia (PPH), and changing feeding schedule has improved regulation in some dogs. This does not appear to be an important cause of poor regulation in cats. Cats tend to eat frequently and acute PPH does not appear to be an important event, especially if the carbohydrate content of the diet is reduced.6 The occasional cat with polyphagia that consumes its ration in one sitting may have acute PPH, which may be confirmed with a BG curve. Dividing the daily ration into more frequent feedings (an automated feeder may help) is one solution.

Problems with Oral Hypoglycemic Agent Therapy

The cat may not be receiving its medication consistently, either because it is difficult to pill or is not eating the medication in its food. This should be reviewed, and the cat may be hospitalized for treatment and effect confirmed with BG measurements.

The drug may be ineffective. Glipizide is used most often, with most cats responding within 3 months.7 Failure to respond after that time suggests there is insufficient functional β-cell mass, and cats failing therapy are usually treated with insulin. There are anecdotal reports of response to other OHAs, but, as with glipizide, failure of response in most cases is probably due to ineffectiveness of the drug. Acarbose and dietary management has resulted in regulation,3 and acarbose has been anecdotally effective in improving regulation in cats receiving other treatments, but it is unlikely to be effective on its own, especially given the nature of PPH in cats. If there is a specific cause of IR, the role of continuing OHA therapy without initiating insulin while addressing the underlying disorder has not been defined.

Problems with Insulin Therapy

Absorption is the crucial factor in determining insulin potency and duration of action, which together determine insulin effect.8 Variability in absorption is an important cause of inconsistent response in humans and is due to injection technique, patient characteristics, and properties of the insulin.9 Under the most controlled experimental conditions the intra-individual variability in insulin pharmacokinetics in humans is 20%, and > 50% in clinical conditions. The situation is likely similar in cats.10

Potential errors in handling the insulin include outdated insulin, inactivation by exposure to excessive heat or light, violent shaking of the insulin bottle, insufficient mixing of the insulin preparation, drawing up the incorrect dose, and dilution errors. In humans, depth of injection, anatomic site, and delay before needle withdrawal affect absorption, but temperature of the insulin and speed of injection do not.7 Injection factors in cats anecdotally associated with poor regulation include unsuccessful injection, rubbing the injection site after injection, and interscapular versus lateral thorax or flank injection.11 Variable injection technique is more likely to result in inconsistent regulation than in persistent hyperglycemia. Owner technique should be reviewed and consistency of injection emphasized. If an insulin administration error is suspected then the cat may be hospitalized to facilitate consistent injection and BG-lowering effect confirmed. Injection problems may be solved in some cases by changing to injection pens or needle-free injection.

Improper Insulin Duration and Dose

Transient Insulin Effect

The duration of insulin effect is not sufficiently long to achieve adequate BG control, which has been considered to be important in cats.8 A transient effect may be due to rapid absorption and/or to a counterregulatory phenomenon. A transient effect can only be definitively diagnosed and distinguished from the Somogyi effect on the basis of a BG curve. The treatment is either to change to a longer-acting preparation or to give insulin twice a day. Increasing the dose will increase the duration of insulin effect, but carries an unacceptable risk for hypoglycemia. When changing from a shorter to longer-acting insulin preparation, the dose may be increased by 10-25% to account for the reduced potency of the longer-acting preparation, but this is difficult with the small doses used in cats. When changing from once to twice daily insulin injections, the dose should be reduced by > 25% to account for potential overlap in insulin effect. Cats are best regulated with twice daily insulin, so this should be discussed with owners at the onset of treatment.

Delayed insulin effect

Absorption of insulin is normally incomplete, and is more so in cats than in dogs or humans.8 Zinc and protamine-based longer-acting preparations are less completely absorbed than shorter-acting ones. Absorption may be sufficiently incomplete that normoglycemia is not achieved with standard doses (mimicking IR). This was primarily a problem with ultralente. In other cases absorption is delayed such that a good part of the day is required before BG falls. A BG curve demonstrates late-onset reduction in BG or no effect. The recommended treatment is to change from PZI to NPH or pork lente, or, paradoxically, to glargine (different mechanism of absorption). Increasing the dose is not recommended because this increases the risk of late-onset hypoglycemia.

Prolonged Insulin Effect

The insulin effect is sufficiently long so that an overlap occurs resulting in a Somogyi effect. This is most likely to occur with twice daily injection of a long-acting preparation. Treatment options include changing to once daily injection of the existing preparation or to twice daily injection of a shorter-acting preparation.

Too Low a Dose

The peak insulin effect does not result in an adequate BG nadir nor sufficient reduction in BG for the duration of insulin action. BG curve measurements obtained every 2 hours are usually diagnostically adequate. The insulin dose should be increased by 0.25-0.5 U, unless previous dose adjustments have demonstrated that larger increments are required. Usually a minimum of 7 days should be allowed for equilibration to the new dose before re-evaluation.

Too High a Dose (Somogyi Effect)

Too high a dose of insulin results in hypoglycemia followed by rebound hyperglycemia due to counterregulatory hormone secretion. In one series, hypoglycemia developed within 4-8 hours of insulin injection and rebound hyperglycemia (>17 mmol/L [300 mg/dl]) within 3-6 hours of the hypoglycemic nadir.12 Signs of hypoglycemia may be seen--most likely lethargy, weakness, and glazed expression. Somogyi effects may result in either persistent hyperglycemia or inconsistent regulation. The latter occurs because the counterregulatory hormones have a hyperglycemic effect for 1-3 days after the overswing. During this period the excessive insulin dose results in good glycemic control. When the counterregulatory effect is over, the insulin dose causes hypoglycemia. A Somogyi effect re-occurs, and the cycle begins again. Diagnosis requires a BG curve. Hourly measurements are recommended during the period of the anticipated BG nadir. It may be necessary to obtain consecutive BG curves in order to catch the day in which the overswing occurs. Because this is labor-intensive and expensive, another approach is to reduce the dose of insulin and observe response.

Inconsistent Absorption

Insulin absorption is normally inconsistent to a certain extent.8-10,13 If this variation is excessive it will result in inconsistent regulation. This is most likely to occur with a long-acting preparation. Daily BG curves will confirm inconsistent regulation, but it is difficult to prove that inconsistent absorption is the cause because insulin sensitivity may also vary daily.10,14 The most practical approach is to change insulin preparation and observe response.

Insulin Resistance

The term "insulin resistance (IR)" is used in two ways. First, it refers to peripheral IR, which is reduced sensitivity of fat and muscle to insulin due to receptor and post-receptor defects. This is a key pathophysiologic event in type 2 diabetes and is present in most diabetic cats. Second, IR refers to therapeutic IR, where high doses of insulin are needed to regulate a diabetic patient. This is the focus of the remaining discussion. Therapeutic IR may be suspected when a dose > 1.5 units/kg per injection is having a minimum effect on hyperglycemia, or when a cat is well-regulated but requires > 2 units/kg/injection.15 This definition is imprecise and does not take into account potency of the insulin preparation. Some cats with no apparent concurrent disease require higher doses of insulin to be well-regulated, presumably due to incomplete absorption. One approach in this case is to continue therapy unchanged and not perform a potentially fruitless diagnostic search. A disadvantage of this approach is the opportunity may be missed to reverse diabetes if it is secondary. If there are increasing insulin requirements a concurrent/underlying disorder should definitely be pursued.

Anti-insulin antibodies formed in humans treated with animal-source insulin, especially with the less pure older preparations. Antibodies may bind insulin in the subcutaneous depot or in the circulation, and, depending upon affinity, may have no effect, may result in inconsistent response, or may cause insulin resistance.13 Antibody formation in cats is probably of minimal significance.16-17 Nonetheless, if a cat is poorly regulated and it appears to be an insulin-related problem, then a change in insulin is required, which could be a change in insulin species of origin.

Concurrent Diseases

Concurrent diseases cause IR by variably causing receptor defects, post-receptor defects, and secretion of counterregulatory hormones. This may result in either persistent or fluctuating IR and cause and effect with the diabetes may not be clear. If a concurrent disease is identified, it should be noted that: 1) it may not be the (sole) cause of difficult regulation; 2) curative treatment may not be possible or well-defined; 3) successful treatment may not resolve receptor or post-receptor defects.

Diabetes increases the risk for bacterial or fungal infection, which in turn causes IR. A common site of infection is the urinary tract.18-21 Urinalysis may not be diagnostic, so urine culture should be performed early in the investigation. Many diabetic cats will have concurrent periodontal disease. Dentistry should be considered and the effect on regulation noted. If a cause of IR has not been identified during work-up, a therapeutic trial with antibiotics may be considered to treat sub-clinical infection (e.g., pyelonephritis).

Diabetic cats may have chronic pancreatitis or exocrine pancreatic neoplasia.22 Pancreatic lipase immunoreactivity and ultrasound examination are currently the best non-invasive tests for pancreatitis.23 Unfortunately, the treatment of chronic pancreatitis is not well-defined, but prednisolone has been recommended.11 It is not known if excision of pancreatic tumors will improve regulation--in one study concurrent pancreatic neoplasia had no effect on suvival.23 Diabetic cats may also have hepatic lipidosis and cholangiohepatitis. Ultrasound may identify hepatic abnormalities; definitive diagnosis is by liver biopsy. Concurrent liver diseases should be treated using standard recommendations. Nutritional support is important to facilitate insulin therapy by minimizing the risk for hypoglycemia. Pancreatitis, hepatic lipidosis and cholangiohepatitis may be risk factors for ketoacidosis.24

Inflammatory and infectious disorders unrelated to diabetes may occur. The former includes asthma and inflammatory bowel disease, which should be confirmed by standard work-up. Poorly regulated cats at risk should be tested for feline leukemia virus and feline immunodeficiency virus infections.

Sub-clinical cardiomyopathy is an uncommon cause of poor regulation. Echocardiography may be considered if other causes of poor regulation have been ruled-out and/or there are signs of heart disease. Diabetes may result in hypertension,25 which is a cause of cardiac hypertrophy. However, hypertension is usually mild and if heart disease is diagnosed, it is probably primary. Other concurrent diseases are those common to mature and geriatric cats. Hyperthyroidism and chronic renal failure are most common; neoplasia may also occur. Hyperthyroidism causes insulin resistance, but secondary diabetes is rare and the need for very high doses of insulin is uncommon. Treatment of hyperthyroidism does, however, reduce insulin doses and improves consistency of regulation.15 Fructosamine levels are lowered in hyperthyroid cats and this should be taken into account when using this value to assess regulation.26-27

Endocrine Tumors Secreting Counterregulatory Hormones

Diabetes is secondary and may resolve with definitive treatment of the underlying disorder. If such treatment is refused, an option is to regulate with high doses of insulin.

Acromegaly (Excessive Growth Hormone)

Clinical signs in addition to those of diabetes include weight gain, facial dysmorphism, thickened neck and paws, respiratory signs due to pharyngeal thickening, organomegaly, arthropathy and central nervous system signs due to the pituitary mass. Polyphagia may be severe and may precede signs of overt diabetes. Diagnosis relies on measuring elevated growth hormone level (which may be difficult to obtain and is occasionally normal) and/or measurement of elevated insulin like-growth factor 1 (IGF-1), identifying a pituitary mass by intracranial imaging, and ruling-out hyperadrenocorticism.28-32 Recent reports suggest that acromegaly may be underdiagnosed.28-32 Measurement of IGF-1 is recommended in any diabetic cat with inconsistent regulation not attributable to another disorder, and any cat with IR, especially if it is male and/or is demonstrating weight gain and polyphagia. A normal value does not rule-out the disorder. Treatment options include hypophysectomy, cryohypophysectomy and radiation therapy.33-37 If hypophysectomy is performed, histology demonstrating an acidophil adenoma is supportive and immunohistochemistry demonstrating growth hormone is definitive. Insulin dose reduction and close monitoring of BG are important as there can be an abrupt decrease in insulin requirements. L-deprenyl was not effective in one case.38

Hyperadrenocorticism (Excessive Cortisol and/or Sex Hormones)

Pituitary and less commonly adrenal tumors occur. Clinical signs in addition to those of diabetes include fragile skin, alopecia, and pendulous abdomen. Appetite varies. Approach to evaluation of the HPA axis and imaging for pituitary or adrenal masses is similar to dogs.39-40 The disorder may be a feature of multiple endocrine neoplasia and concurrent acromegaly and pituitary-dependent hyperadrenocorticism has been reported.41-42 Effective treatments include trilostane and adrenalectomy; hypophysectomy, cryohypophysectomy, and irradiation are alternatives for pituitary tumors.36,39-40,43 If hypophysectomy is performed, histology demonstrating a chromophobe adenoma is supportive and immunohistochemistry demonstrating ACTH is definitive. As with acromegaly, abrupt decreases in insulin requirements may occur following treatment.

Pheochromocytoma (Excessive Epinephrine/Norepinephrine)

This is a rare tumor in cats that may cause PU/PD, but associated IR diabetes has not been reported.

Glucagonoma (Excessive Glucagon)

This pancreatic tumor has not been reported in cats.

Drugs Causing Insulin Resistance

A number of drugs affect carbohydrate, fat and protein metabolism, and counterregulatory hormones, with variable clinical implications, and these should be considered during their use. For example, metoclopramide may increase growth hormone levels. With respect to diabetogenic drugs per se, a diabetic cat may have a concurrent disorder necessitating glucocorticoid therapy (e.g., asthma). Glucocorticoids are not contraindicated under these circumstances, but insulin doses will have to be increased. There is little justification for using progestins for behavior problems because other behavior altering drugs are available.


1.  Ristic JM, et al. J Feline Med Surg 2005;7(3):153.

2.  Wiedmeyer CE, et al. Diabetes Technol Ther 2005;7(6):885.

3.  Mazzaferro EM, et al. J Feline Med Surg 2003;5(3):183.

4.  Bennett N, et al. J Feline Med Surg 2006;8(2):73.

5.  Kirk CA. Vet Clin North Am Small Anim Pract 2006;36(6):1297.

6.  Martin GJW, et al. J Feline Med Surg 1999;1(4):241.

7.  Feldman EC, et al. J Am Vet Med Assoc 1997;210(6):772.

8.  Abrams-Ogg ACG. Proceedings 25th ACVIM Forum 2007:271.

9.  Gin H, et al. Diabetes Metab 2005;31(1):7.

10. Alt N, et al. J Am Vet Med Assoc 2007;230(7):1011.

11. Fondacaro JV, et al. in August JR (ed). Consultations in Feline Internal Medicine 4. Saunders, 2001:175.

12. McMillan FD, et al. J Am Vet Med Assoc 1986;188(12):1426.

13. Nolte MS. Endocrinol Metab Clin North Am 1992;21(2):281.

14. Feldhahn JR, et al. J Feline Med Surg 1999;1(4):233.

15. Peterson ME. Vet Clin North Am Small Anim Pract 1995;25(3):691.

16. Harb-Hauser M, et al. J Vet Intern Med 1998;12(3):245.

17. Hoenig M, et al. Vet Immunol Immunopathol 2000;77(1):93.

18. Kirsch M. Tierärztl Prax 1998;26(1):32.

19. Jin Y, et al. J Am Animal Hosp Assoc 2005;41(6):373.

20. Bailiff NL, et al. J Vet Intern Med 2006;20(4):850.

21. Mayer-Roenne B, et al. J Feline Med Surg 2007;9(2):124.

22. Goossens MM, et al. J Vet Intern Med 1998;12(1):1.

23. Forman MA, et al. J Vet Intern Med 2004;18(6):807.

24. Bruskiewicz KA, et al. J Am Vet Med Assoc 1997;211(2):188.

25. Sander S, et al. Tierärztl Prax 1998;26(2):110.

26. Graham PA, et al. Res Vet Sci 1999;67(2):171.

27. Reusch CE, et al. J Am Vet Med Assoc 1999;215(9):1297.

28. Niessen SJ, et al. Vet Rec 2007;160(26):902.

29. Niessen SJ, et al. J Vet Intern Med 2007;21(5):899.

30. Berg RI, et al. J Vet Intern Med 2007;21(5):892.

31. Alt N, et al. Res Vet Sci 2007;83(3):331.

32. Reusch CE, et al. Vet Rec 2006;8(5):309.

33. Abrams-Ogg ACG, et al. Can Vet J 1993;34(11):682.

34. Blois S, et al. J Small Anim Pract, submitted.

35. Littler RM, et al. J Small Anim Pract 2006;47(7):392.

36. Mayer MN, et al. J Vet Intern Med 2006;20(5):1151.

37. Goossens MM, et al. J Am Vet Med Assoc 1998;213(3):74.

38. Abraham KA, et al. Aust Vet J 2002;80(8):479.

39. Gunn-Moore D. Vet Clin North Am Small Anim Pract 2005;35(1):171.

40. Chiaramonte D, et al. Clin Tech Small Anim Pract 2007;22(1):26.

41. Rocca-bianca P, et al. Vet Pathol 206;43(3):345.

42. Meij BP, et al. J Comp Pathol 2004;130(2-3):209.

43. Meij BP. Vet Clin North Am Small Anim Pract 2001;31(5):1015.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Anthony Abrams-Ogg, DVM, DVSc, DACVIM(SAIM)
University of Guelph
Guelph, ON, Canada

MAIN : ACVIM SAIM Generalist : Difficult-to-Regulate Diabetic Cats
Powered By VIN