Identifying the reasons behind difficult-to-control diabetes in dogs

Diabetes mellitus is a common endocrinopathy in canine practice, but successful patient management sometimes can be a challenge. Some dogs appear to respond inadequately to a standard dose of insulin or have dramatic swings from hypoglycemia to hyperglycemia. This article reviews reasons for poor diabetic regulation and outlines a logical approach to the difficult-to-treat diabetic dog.

APPROACH TO THE PROBLEM DIABETIC DOG

In general, problems with diabetic control can be categorized as insulin-related, client-related, or patient-related (Table 1). Although patient-related problems are more common, it is wise to carefully exclude insulin- and client-related issues first, since these are often easily identified and addressed. When insulin- and client-related issues have been ruled out, we then start to look for patient-related problems.

Table 1

Insulin-related problems

Sudden loss of regulation in a previously well-controlled diabetic may occur because of problems with the insulin itself. Even if you have no specific reason to suspect the insulin has decreased biologic activity, it is always wise to discard the present bottle and start a new one. Insulin is a peptide and, therefore, can be damaged by exposure to heat or extreme cold. With some insulin types, agitation during shipping can also damage the molecule and alter its biologic effects. Another consideration is bacterial contamination; this can occur quickly and result in degradation of the insulin molecules and loss of potency. And lastly, dilution of the insulin can cause problems as the product may have altered activity or may have become unstable. Before conducting an exhaustive and expensive search to identify patient-related problems leading to insulin resistance, it may be worthwhile to replace an older insulin vial with a new, undiluted one.

Client-related problems

Client-related problems are most likely to occur relative to dogs with newly diagnosed diabetes. Even with careful instruction, many clients struggle to draw up the correct insulin dose or have problems administering it. Clients often have difficulty identifying and expelling air bubbles and inadvertently underdose their pets. Using the smallest possible insulin syringe may avoid some of these issues, as the larger scale makes it easier to see air bubbles and other problems. In addition, insulin syringe magnifiers can be purchased from pharmacies and may facilitate accurate dosing. Another consideration is confusion about U40 and U100 syringes. If a U40 insulin is administered with a U100 syringe, the actual dose given is 40% of the anticipated amount. This problem is commonly encountered when clients change insulin types and do not discard the old syringes, or if they choose to use their own medical supplies.

It is often helpful to watch the client draw up and administer an insulin dose. Watch carefully for problems with the injection technique, such as intradermal injection or a through-and-through injection. Some owners tend to withdraw the syringe while injecting, so the insulin ends up lost in the pet's coat. Checking for moisture on the hair after the injection can be helpful, as can experimenting with a different needle size or length. Practicing with 0.9% saline solution injections is a safe and effective way to train clients and increase their confidence. Shaving a small patch of hair lets clients visualize the needle penetrating the skin and may improve their technique.

Patient-related problems

When insulin- and client-related problems have been ruled out, a careful hunt for problems with the patient becomes necessary. Patient-related problems are often the most challenging to identify, and a logical approach is essential. In general, patient-related issues reflect changes in energy use or intake or problems with insulin absorption, kinetics, or resistance. The clinical manifestations of these problems are variable; some dogs experience unpredictable fluctuations in their blood glucose concentrations (i.e. poor regulation), while others require excessive amounts of insulin (i.e. insulin resistance).

Although several monitoring methods may be considered in clinically stable diabetic dogs, the key to identifying and resolving patient-related problems is frequent blood glucose measurements. The best way to collect an accurate picture of the patient's glycemic control is with serial blood glucose measurements, often referred to as a glucose curve. Glucose curves may show some day-to-day variation, so it is important to correlate clinical signs with blood glucose measurements.1 With some patients, it may be worthwhile to start an at-home monitoring program so sufficient representative data can be collected quickly. After proper instruction, most clients can measure their dogs' blood glucose concentrations at home. This is more cost effective for the client and often much easier on the dog. A measurement should be obtained every two hours throughout the day, starting before insulin is administered and food is offered. It is not adequate to simply perform a few spot checks over the course of a day. The reliability of the handheld glucometers appears to be quite variable, and a veterinary-validated system is probably the best option. In a recent study comparing several different devices, the AlphaTRAK (Abbott Animal Health) had the lowest percentage of misclassified results at 2.1%, compared with 4.2% to 38.7% for the other systems evaluated.2 An alternative option is a continuous blood glucose monitor.3 Several different models are on the market, but the Guardian REAL-Time Continuous Glucose Monitoring System (Medtronic) has been evaluated most frequently in veterinary species. In general, these systems consist of a small probe that sits in the interstitial space for a few days and transmits the blood glucose reading to a small receiver. The receiver has to stay fairly close to the patient but is easily attached to a harness or collar. Daily calibration is necessary, but these systems can be helpful in poorly regulated patients.

Poor regulation. Poorly regulated patients require frequent adjustments to their insulin dose or experience episodes of unexplained hypoglycemia (Table 2). Some poorly regulated patients respond bizarrely to insulin, with dramatic fluctuations in blood glucose concentrations throughout the day, or have "upside-down" curves, in which the blood glucose concentration rises after insulin administration. In other cases, the clinic data do not correlate well with the data obtained at home, making dose adjustment decisions increasingly difficult.

Table 2

In general, poor regulation reflects problems with energy use or intake or the biological behavior of the insulin itself. As energy issues are easily identified and addressed, it is helpful to start with this possibility before moving forward.

In most cases, carefully questioning the client will highlight issues with energy utilization. Any sporadic exertion can substantially affect calorie utilization and may result in profound hypoglycemia. It's important to ask about recent changes in routine, such as a visit from grandchildren or a trip to the groomer. We may not think of it as exercise, but a day spent circling in a cage at the doggy spa has the same physiologic effects as a run in the park.

Changes in energy intake may be more difficult to identify. Some clients hesitate to admit breaking the dietary rules or feel that the occasional snack or shared sandwich will not cause problems. It is important to explain that a couple of cookies or a piece of buttered toast can provide 20% of the daily caloric needs for a small patient and can certainly impact diabetic regulation. Make sure every member of the household is onboard with the feeding plan and understands the importance of a consistent diet. The dog should be fed on a regular schedule, accompanied by daily moderate exercise.

When energy issues have been ruled out, problems with how the patient interacts with the insulin should be considered next. Problems with erratic insulin absorption are relatively uncommon. However, some animals develop thickened skin and irritation of the subcutaneous tissues if the insulin is always administered at the same site. This is a consideration in dogs with a variable or delayed onset of insulin action, often manifested by the inverted glucose curve. With these patients, serial blood glucose concentrations are markedly different if a new injection site is used. Simply teaching the owner to rotate through six or eight sites on the flanks and dorsum usually solves this problem.

Insulin kinetics problems include an unexpectedly short or long duration of effect. These patients respond to the insulin, with an appropriate decrease in their blood glucose concentration, but the duration of control is unsatisfactory. These problems are often most easily identified with 24-hour serial blood glucose measurements or with a continuous blood glucose monitor.

Although most dogs receiving Lente or NPH insulin products require twice-daily dosing, some dogs experience a longer duration of effect and are prone to hypoglycemia with twice-daily administration (Table 3).4,5 These dogs may need once-daily dosing with a combination of insulin products to achieve acceptable control. An engineered long-acting product such as insulin detemir or glargine may be a useful option for these patients. Both insulins are designed for background glycemic control in people. In this scenario, the human patient simply administers a dose of regular insulin with each meal and relies on the background product for sustained glycemic control. These insulins have a slow onset but prolonged action and can be effectively incorporated into canine protocols, often on a once-daily basis.6 It is important to remember that these insulins are pH-balanced and cannot be mixed in the same syringe with other insulins.

Table 3

Conversely, both Lente and NPH may only last six to eight hours in some dogs, and three-times-daily dosing or the addition of a longer-acting insulin may be necessary. Again, insulin detemir or glargine can be useful in these cases. An example of a dog requiring two insulin types is provided in Table 4. Consultation with a specialist about starting doses may be helpful before beginning the use of an unfamiliar insulin.

Table 4

Historically, anti-insulin antibodies have been implicated in problems with diabetic regulation. It has been suggested that antibodies may cause trapping of insulin at the injection site with subsequent erratic release from the subcutaneous tissues or overt insulin resistance (see below).7 However, the clinical impact of anti-insulin antibodies in dogs is still unclear. Recent studies have indicated that most dogs receiving heterologous exogenous insulins produce anti-insulin antibodies, although this was not associated with poor diabetic control.8,9 The only licensed insulin product for dogs in the United States is of porcine origin; porcine-origin insulin is identical to canine insulin and is not expected to trigger anti-insulin antibody production. Human insulin is highly analogous to the canine molecule and is also likely to be well-tolerated. Beef-origin insulin is much more likely to trigger antibodies in dogs. At the present time, the only beef insulin available in the United States is a PZI formulation. It is inherently a poor choice for dogs because of its biologic behavior, and it should probably be avoided for this reason alone. Patients with clinically important anti-insulin antibodies would be expected to respond appropriately to intravenous regular insulin or a switch to a homologous product.

Some drugs (e.g. beta blockers) and other diseases (e.g. hypoadrenocorticism, insulin-secreting tumors) can result in unexpected hypoglycemic episodes in diabetic dogs.10-12 A careful check of concurrent medications and a full general health evaluation are necessary to identify these uncommon causes of poor diabetic regulation.

Insulin resistance. No firm rules dictate when to classify a dog as insulin resistant, but it is generally appropriate to consider a patient to have poor glycemic control when its glucose concentrations are > 300 mg/dl despite insulin doses > 1.5 U/kg/dose.7

Insulin-resistant dogs often show a poor response to intravenous insulin, as the problem is due to impaired interaction with the insulin receptor or altered post-receptor translational events. We tend to think of insulin as working like a key in a lock to open up glucose transport into cells, but its action is far more complex than that. Insulin profoundly impacts energy storage in fat and muscle, as well as simply lowering blood glucose concentrations. Consequently, numerous metabolic events can impact the action of insulin (Table 2).13 The classic causes of canine insulin resistance are described in the following section. See Table 5 for a suggested approach to these cases.

Table 5

The impact of obesity on insulin sensitivity is well-understood in human medicine and is becoming recognized in the veterinary field.14-16 Although many diabetics lose weight as a result of the disease, some are still obese when treatment is initiated. Obesity may blunt the response to exogenous insulin and make regulation more difficult. An appropriate weight loss program should be started immediately, with the goal of reaching an optimal body condition score within four months. In most cases, switching to fixed portions of a low-fat, high-fiber food will be helpful. The new weight loss drug for dogs (dirlotapide; Slentrol—Pfizer Animal Health) is probably not an appropriate choice for weight loss in diabetic patients because food intake can fluctuate daily and vomiting is a common side effect.17

Hyperlipidemia is associated with insulin resistance in diabetic people. Experimental studies also support a link in dogs, but little clinical data addresses this issue in spontaneously diabetic dogs.18 However, a fasting serum triglyceride concentration > 800 mg/dl should be addressed in an insulin-resistant diabetic patient. Possible causes include inherited lipid disorders and hypothyroidism. Switching to a markedly fat-restricted diet is often effective. Additional interventions may include omega-3 fatty acid supplementation or lipid-lowering agents such as niacin or gemfibrozil.19

Concurrent infection is probably the number one reason for acute insulin resistance. Urinary tract infections are routinely found in diabetic dogs and should be considered in any poorly controlled patient.20 A urine culture is often the first step in the evaluation of an insulin-resistant diabetic, even if the urinalysis is unremarkable and clinical signs of infection are not reported. In one study of dogs with diabetes, less than 5% of those with positive culture results demonstrated stranguria or pollakiuria.21 It is important to remember that glucosuria provides an ideal environment for bacterial colonization, while the diuretic effect of the glucose dilutes the usual sediment examination findings of white blood cells and bacteria.

The other classic spot for an occult infection is the oral cavity.22 A thorough dental examination should be performed, and any suspicious areas should be investigated further. In some cases, intraoral radiographs may be the only way to identify a subclinical tooth root abscess.

Other endocrinopathies should be considered in dogs with insulin resistance. Hypothyroidism is a recognized cause of insulin resistance and was noted in 4% of diabetic dogs in one study.20,23 Patients with insulin resistance due to thyroid disease often slowly gain weight despite persistent hyperglycemia. This is a useful finding, as most unregulated diabetic dogs become thinner over time. Persistent hypercholesterolemia may also be noted on routine laboratory tests, although this can be explained by poor diabetic regulation for any reason. A mild nonregenerative anemia may develop with long-standing hypothyroidism but is not reported in most cases. It is usually necessary to measure total thyroxine (T4), free T4 by equilibrium dialysis, and thyroid stimulating hormone (TSH) concentrations to confirm the diagnosis. A total T4 concentration is often subnormal due to euthyroid sick syndrome in a poorly controlled diabetic and is, therefore, an unreliable indicator of thyroid status. The classic picture for spontaneous hypothyroidism is subnormal T4 and free T4 concentrations, along with increased TSH concentrations. However, it is important to note that the TSH concentration may be normal in a subset of dogs with confirmed primary hypothyroidism.24 If results are equivocal, consultation with an internist may be helpful.

Concurrent hyperadrenocorticism is commonly reported in dogs with diabetes mellitus and results in severe insulin resistance.20,25 Dogs with untreated hyperadrenocorticism and diabetes mellitus often maintain a blood glucose concentration > 400 mg/dl despite large doses of insulin. This is probably one of the more difficult disorders to identify in poorly regulated diabetic dogs, as the clinical signs of hyperadrenocorticism—namely excessive thirst, hunger, and urination—can all be explained by persistent hyperglycemia. Routine laboratory findings associated with hyperadrenocorticism, such as hypercholesterolemia and increased alkaline phosphatase activity, are also noted in diabetic patients, so standard laboratory testing is often of little use. In addition, the stress of poor diabetic regulation can impact adrenal function tests and cause adrenomegaly. Opinions differ about the best screening test for hyperadrenocorticism in diabetic dogs, as results of both the low-dose dexamethasone suppression (LDDS) test and the adrenocorticotropic hormone (ACTH) stimulation test are influenced by concurrent disease.26,27 I personally prefer the LDDS test in these cases but am not aware of any definitive data to support its use over the ACTH stimulation test. A urine cortisol-creatinine ratio would be of little use, as false positive results commonly occur in dogs with nonadrenal disease.28 Additional clinical evidence of hyperadrenocorticism (e.g. poor hair growth, delayed healing) along with additional diagnostics (e.g. abdominal ultrasonography) should be considered before you initiate treatment for hyperadrenocorticism.

Transient acromegaly can occur in intact females during the diestrus phase of the reproductive cycle and will result in insulin resistance. In these cases, excess growth hormone is released from mammary tissue, under the influence of progesterone from the ovaries.29 Any intact female diabetic dog should undergo prompt ovariohysterectomy when its diabetes is stabilized to prevent the predictable insulin resistance associated with diestrus. In addition, some dogs may experience diabetes remission after ovariohysterectomy if enough functional pancreatic beta cells remain. Rare cases of acromegaly secondary to a functional pituitary tumor have been reported, and this also results in substantial insulin resistance.30 Classic physical examination findings for dogs with pituitary acromegaly include thickened skin, redundant skin folds, and an enlarged tongue. Respiratory stridor is common because of growth of the pharyngeal soft tissue. This is an uncommon disorder in dogs compared with cats, and the diagnosis would be based on clinical signs and confirmation of a pituitary mass.

Sterile inflammatory disease can also cause insulin resistance, probably through the release of cortisol. Although any body organ may be affected, pancreatitis is a common problem in diabetic dogs and may be sustained or severe enough to affect diabetic regulation.20 Skillful ultrasonography or the measurement of canine pancreas-specific lipase activity (Spec cPL—IDEXX) may be helpful in cases of suspected pancreatitis.31 No specific therapy exists for this disorder, but pain control and a switch to a strictly low-fat diet are often beneficial.

Another important cause of insulin resistance is concurrent medication. Clearly, any products containing glucocorticoids should be avoided unless absolutely necessary. This includes topical products, such as ear treatments and skin-soothing sprays. It is not unusual for clients to purchase these medications for other pets and then use them with the best intentions on their diabetic animals. Over-the-counter products marketed for people rarely contain enough corticosteroids to cause substantial insulin resistance, but it is always prudent to check. Synthetic progestins, such as megestrol acetate and medroxyprogesterone acetate, are potent insulin antagonists and should not be administered to diabetic dogs.32

SUMMARY

The difficult-to-control diabetic dog can be frustrating, but most patients can be effectively managed if the cause of poor regulation is identified and addressed. Once insulin- and client-related problems have been excluded, patient-related issues should be pursued. The cornerstone of these investigations is reliable serial glucose monitoring, either at home or in the clinic. A stepwise, logical approach should be followed so that straightforward problems are effectively addressed before more complex diagnoses are considered.

Audrey K. Cook, BVM&S, MRCVS, DACVIM, DECVIM

Department of Small Animal Clinical Sciences

College of Veterinary Medicine

Texas A&M University

College Station, TX 77843

REFERENCES

1. Fleeman LM, Rand JS. Evaluation of day-to-day variability of serial blood glucose concentration curves in diabetic dogs. J Am Vet Med Assoc 2003;222:317-321.

2. Cohen TA, Nelson RW, Kass PH, et al. Evaluation of six portable blood glucose meters for measuring blood glucose concentration in dogs. J Am Vet Med Assoc 2009;235(3):276-280.

3. Weidmeyer CE, Johnson PJ, Cohn LA, et al. Evaluation of a continuous glucose monitoring system for use in dogs, cats, and horses. J Am Vet Med Assoc 2003;223:987-992.

4. Fleeman LM, Rand JS, Morton JM. Pharmacokinetics and pharmacodynamics of porcine insulin zinc suspension in eight diabetic dogs. Vet Rec 2009;164(8):232-237.

5. Palm CA, Boston RC, Refsal KR, et al. An investigation of the action of Neutral Protamine Hagedorn human analogue insulin in dogs with naturally occurring diabetes mellitus. J Vet Intern Med 2009;23:50-55.

6. Mori A, Sako T, Lee P, et al. Comparison of time-action profiles of insulin glargine and NPH insulin in normal and diabetic dogs. Vet Res Commun 2008;32:563-573.

7. Feldman EC, Nelson RW. Canine diabetes mellitus. In Feldman EC, Nelson RW, eds. Canine and feline endocrinology and reproduction. 3rd ed. St. Louis, Mo.: Elsevier, 2004;486-538.

8. Davison LJ, Ristic JME, Herrtage ME, et al. Anti-insulin antibodies in dogs with naturally occurring diabetes mellitus. Vet Immunol Immunopathol 2003;91:53-60.

9. Davison LJ, Walding B, Herrtage ME, et al. Anti-insulin antibodies in diabetic dogs before and after treatment with different insulin preparations. J Vet Intern Med 2008;22(6):1317-1325.

10. Plumb DC. Plumb's veterinary drug handbook. 6th ed. Ames, Iowa: Blackwell Publishing, 2008:1048-1051.

11. Lathan P, Tyler J. Canine hypoadrenocorticism: pathogenesis and clinical features. Compend Contin Educ Pract Vet 2005;27:110-120.

12. Bryson ER, Snead ECR, Chantal McMillan C, et al. Insulinoma in a dog with pre-existing insulin-dependent diabetes mellitus. J Am Anim Hosp Assoc 2007;43:65-69.

13. Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000;106:171-176.

14. Kahn BB, Flier JS. Obesity and insulin resistance. J Clin Invest 2000;160:437-481.

15. Gayet C, Bailhache E, Dumon H, et al. Insulin resistance and changes in plasma concentration of TNFalpha, IGF1, and NEFA in dogs during weight gain and obesity. J Anim Physiol Anim Nutr 2004;88:157-165.

16. Ellmerer M, Hamilton-Wessler M, Kim SP, et al. Reduced access to insulin-sensitive tissues in dogs with obesity secondary to increased fat intake. Diabetes 2006;55:1769-1775.

17. Wren JA, Gossellin J, Sunderland SJ. Dirlotapide: a review of its properties and role in the management of obesity in dogs. J Vet Pharmacol Therap 2007;30(suppl 1):11-16.

18. Serisier S, Gayet C, Leray V, et al. Hypertriglyceridaemic insulin-resistant obese dog model: effects of high-fat diet depending on age. J Anim Physiol Anim Nutr 2008;92:419-425.

19. Johnson MC. Hyperlipidemia disorders in dogs. Compend Contin Educ Pract Vet 2005;27:361-371.

20. Hess RS, Saunders M, Van Winkle TJ, et al. Concurrent disorders in dogs with diabetes mellitus: 221 cases (1993-1998). J Am Vet Med Assoc 2000;217:1166-1173.

21. Forrester SD, Troy GC, Dalton MN, et al. Retrospective evaluation of urinary tract infection in 42 dogs with hyperadrenocorticism or diabetes mellitus or both. J Vet Intern Med 1999;13(6):557-560.

22. Van Nice, E. Management of multiple dental infections in a dog with diabetes mellitus. J Vet Dent 2006;23:18-25.

23. Ford SL, Nelson RW, Feldman EC, et al. Insulin resistance in three dogs with hypothyroidism and diabetes mellitus. J Am Vet Med Assoc 1993;202:1478-1480.

24. Ferguson DC. Testing for hypothyroidism in dogs. Vet Clin North Am Small Anim Pract 2007;37:647-669.

25. Hume DZ, Drobatz KJ, Hess RS. Outcome of dogs with diabetic ketoacidosis: 127 dogs (1993-2003). J Vet Intern Med 2006;20:547-555.

26. Geiger TL, Feldman EC, Wallack ST, et al. Lymphoma as a model for chronic illness: effects on adrenocortical function testing. J Vet Intern Med 2003;17:154-157.

27. Behrend EN, Kemppainen RJ, Clark TP, et al. Diagnosis of hyperadrenocorticism in dogs: a survey of internists and dermatologists. J Am Vet Med Assoc 2002;220:1643-1649.

28. Smiley LE, Peterson ME. Evaluation of a urine cortisol:creatinine ratio as a screening test for hyperadrenocorticism in dogs. J Vet Intern Med 1993;7:163-168.

29. Rijnberk A, Mol JA. Progestin-induced hypersecretion of growth hormone: an introductory review. J Reprod Fertil 1997;51(Suppl 1):335-338.

30. Fracassi F, Gandini G, Diana A, et al. Acromegaly due to a somatroph adenoma in a dog. Domest Anim Endocrinol 2007;32:43-54.

31. Xenoulis PG, Suchodolski JS, Steiner JM. Chronic pancreatitis in dogs. Compend Contin Educ Pract Vet 2008;30:166-180.

32. Plumb DC. Plumb's veterinary drug handbook. 6th ed. Ames, Iowa: Blackwell Publishing, 2008:760-765.