Feline diabetes mellitus (Proceedings)

Pathophysiology

Diabetes mellitus (DM) can be defined as a group of metabolic diseases characterized by hyperglycemia, which results from defects in insulin secretion, insulin action, or both. DM can occur because of disease affecting the endocrine pancreas and other endocrinopathies such as acromegaly.

Diabetes mellitus in cats is one of the most frequently identified endocrinopathies in cats with incidence rates of 1 in 100 to 1in 400 reported (Panciera et.al. 1990, Rand JS et.al. 1997). Based upon islet histopathology, risk factors, and clinic course of disease, cats seem to develop a form of DM that is most closely related to type-2 DM in humans. Type-1 DM is caused by immune-mediated destruction of beta cells by antibody and T cells, which seems to be a much less common occurrence in cats. Type-2 diabetes is characterized by beta cell dysfunction leading to inadequate insulin secretion and impaired insulin action or insulin resistance.

Insulin resistance is a condition in which a "normal" amount of secreted insulin produces a subnormal response. In an insulin resistant state more insulin has to be secreted to produce the same glucose lowering effect as compared to a "normal" amount. Diabetic cats have been found to be about six times less sensitive to insulin than healthy cats (Feldhahan JR, et.al.1999). In humans there is evidence that inheritance plays a large role in insulin resistance, but it is increased by environmental or lifestyle factors. This is likely similar in cats. In support of there being a genetic bases of disease in cats, in Australia Burmese cats are over-represented accounting for approximately 20% of diabetic cats (Rand JS et.al. 1997). Obesity in cats as in humans seems to play a role in the development of the insulin resistant state. In one study cats that were allowed to become obese demonstrated a 52% decrease in tissue sensitivity to insulin. Also normal weight cats with initial lower insulin sensitivities had a greater risk of developing impaired glucose tolerance as they became obese (Appleton DJ et.al. 2001). Persistent decreased insulin sensitivity/insulin resistance results in hyperglycemia, which in turn leads to excessive production of insulin and beta cell dysfunction and exhaustion. In humans and rats, exercise has been shown to increase insulin sensitivity and lack of exercise impairs insulin action and can add to insulin resistance. Not all cats with DM are or have been obese, however. Diet may also be a factor in the development of DM in cats. Many commercial cat foods contain high levels of highly digestible carbohydrates. Diets high in refined carbohydrates can produce a greater postprandial increase in glucose and insulin secretion. In cats it is thought that these high carbohydrate diets lead to life-long increased demand for insulin secretion and resultant beta cell exhaustion.

Beta-cell dysfunction in humans occurs in stages as beta cell failure progresses: glucose desensitization, beta-cell exhaustion, and glucose toxicity. Some of these stages have been identified in cats. In health insulin secretion occurs in two phases. This first phase is a rapid release of insulin from preformed granules in the beta-cell occurring within a few minutes of ingestion, and the second phase after then first and lasts until normoglycemia is restored. In the affected cat, in response to an intravenous glucose bolus, the first phase of insulin secretion is markedly reduced and the second phase is delayed and exaggerated. As glucose intolerance progresses the second phase of insulin secretion is lost: beta-cell exhaustion. With persistent hyperglycemia glucose toxicity occurs. The mechanism by which glucose toxicity suppresses insulin secretion is unclear. Glucose toxicity leads to irreversible changes that permanently affect the cells ability to produce insulin. The cells with a functional reserve try to compensate and may become hyperfunctioning. It is thought that this increases their susceptibility to glucose toxicity and increases the likelihood of permanent DM (permanent insulin-dependent state). Sulphonylureas, oral hypoglycemic drugs, act by stimulating beta cells to increase insulin secretion and as such may actually increase exhaustion of remaining beta cells. Histopathologic changes seen in glucose toxicity include glycogen deposition and cell death. The degree of severity of glucose toxicity depends on the degree and duration of hyperglycemia (Link KRJ, et.al. 1996).

The most consistent histopathologic finding in cats with DM is islet amyloidosis. More than 80% of diabetic cats have been shown to have islet amyloidosis; although not all cats with islet amyloidosis have DM. Amyloid is derived from the hormone amylin, which is secreted with insulin from beta cells. Amyloid fibrils have been shown to be toxic to beta cells, Amyloid deposition likely contributes to beta cell loss, insulin deficiency, and the progression of DM in cats.

Transient diabetes or diabetic remission is reported to occur 20 up to 84% of adequately treated cats (Feldman and Nelson 2004; Roomp K, et, al. 2009). These cats present with clinical signs and laboratory findings of DM, but clinical signs and blood glucose levels return to normal weeks to a few months after initiating treatment and achieving good diabetic control. Remission may last months to years but may be only weeks in some cats. It is suspected that these cats have subclinical disease and become clinical when illness places extra demands on the pancreas (increased insulin resistance) or when given drugs like glucocorticoids or progesterone or prior to the onset of clinical diabetes. It seems that remission is more likely with early treatment and intensive monitoring to obtain good glycemic control (Roomp K et.al. 2009).

Diagnosis

Diagnosis of diabetes mellitus in cats is based upon consistent clinical signs, persistent hyperglycemia, and glucosuria. Diabetic cats commonly exhibit polyuria, polydipsia, polyphagia, weight loss, and muscle wasting. Some diabetic cats may be obese. Some cats may have an unkempt hair coat, lethargy, and weakness. Diabetic neuropathy occurs in approximately 10% of cats (Feldman and Nelson 2004); cause is unknown but may involve abnormalities in the sorbitol metabolic pathway. Neurologic signs can range from hind limb weakness and impaired jumping ability to, most evident, a plantigrade stance. Cataracts are rare in cats. Life threatening complications seen in human diabetics, diabetic nephropathy, vasculopathies, coronary artery disease, require a very long time to develop and are not known to occur in cats. Abnormalities seen on complete blood count are non-specific for DM. The most commonly seen biochemical abnormalities in the diabetic cat include hyperglycemia, elevated ALP and ALT activities, and hypercholesterolemia. Glucosuria is also present. Transient stress hyperglycemia is a common problem in cats and can cause serum glucose to increase above 300 mg/dL. Glucosuria can sometimes develop in these cats as well although there is usually just trace or 1+ glucose seen on urine dipsticks. If there is concern that stress is causing hyperglycemia/glucosuria owners may be sent home with urine-dipsticks to monitor for glucosuria in a non-stressful environment or alternatively a fructosamine can be measured. Serum fructosamine concentration reflects average blood glucose level over 1-3 weeks. Urine culture should be performed in all cats diagnosed with DM. Positive urine cultures were reported to occur in 12% of cats with DM surveyed, many of these cats had no signs of lower urinary tract disease nor urine sedimentation that was consistent with a urinary tract infection (Mayer-Roenne B, et.al. 2007).

Treatment

The goals of diabetes mellitus treatment are to correct the major clinical signs of diabetes (weight loss, PU, PD, polyphagia, inappetence, etc), and to prevent diabetic complications (DKA, hypoglycemia, neuropathy, etc). With good glycemic control through appropriate therapy diabetic remission is possible for some cats, but it is not possible to predict which cats will have a diabetic remission and become non-insulin dependent.

Dietary considerations: High/moderate protein, low carbohydrate diet with multiple meals (4) fed throughout the day. Ad libitum feeding should be avoided unless a cat is able to maintain adequate body condition with this type of feeding. Traditionally high fiber diets have been recommended for feeding diabetic cats. Fiber can blunt postprandial hyperglycemia by decreasing intestinal carbohydrate absorption. One study demonstrated that a high fiber diet was more beneficial to some cats than a low fiber diet. However the low fiber diet had increased carbohydrate levels than the low fiber diet (Nelson RW et.al 1994). Cats in the wild eat approximately 12-20 small meals spread throughout the day and night with the main energy source being protein. Cats have a gastrointestinal tract and metabolic pathways that are geared away from carbohydrate and toward obtaining energy from a high protein diet. Cats have less disaccharidase activity and no hepatic glucokinase activity. Hepatic glucokinase is responsible for glucose phosphorylation in the liver to trap glucose in hepatocytes for glucose oxidation. Their high protein diet supplies amino acids that are used in gluconeogenesis for energy and continually produce glucose. Cats on a high protein diet as in the natural setting do not experience postprandial hyperglycemia as their glucose production from gluconeogenesis is continual. However, when exposed to a high carbohydrate diet, they are less able for reasons described above to adapt to the postprandial glucose surge and have significant increases in blood glucose concentration. One study compared a low carbohydrate, low fiber diet and moderate carbohydrate, high fiber diet for feeding of diabetic cats in addition to insulin therapy. Cats in both groups had diabetic remission; however there was a significant difference in number of cats achieving remission (68% low carb vs 41% mod carb). Both diets had very similar protein concentrations (Bennett N et.al. 2006). Weight loss is an essential component to management of DM in obese cats. A reasonable target weight should be identified and cats fed approximately 70-80% maintenance requirements for that targeted weight. Weight loss should be very slow and gradual (approximately 1-2% body weight per week). It is important to remember that as weight decreases, insulin requirements will likely also decrease.

The α-glucosidase inhibitors (eg acarbose) reduce intestinal glucose absorption (Greco 1999) and are generally not effective in the treatment of feline diabetes alone, but can be used in conjunction with insulin and/or other oral agents to gain better glycemic control. This may be useful in cats with concurrent chronic kidney disease that are fed protein-restricted diets.

Insulin therapy is the mainstay for treatment of feline DM. Long-term insulins that are available for use in cats include glargine insulin, protamine zinc insulin (PZI), human-recombinant neutral protamine hagedorn (NPH) insulin, and insulin detemir. Feline insulin is most similar in amino acid sequence to bovine insulin, differing by only one amino acid. Feline insulin differs from pork-based insulin by 3 amino acids and from human insulin by 4 amino acids. However at this time animal based insulins are only available through veterinary compounding pharmacies. Human recombinant or human insulin analogs fortunately work well for diabetic management of cats.

NPH is considered an intermediate acting insulin and comes in a concentration of 100 U/mL (U 100). This insulin may have too short of a duration of action for many cats if used as a twice daily insulin, but can be used three times daily, owner schedule permitting.

PZI insulin is considered a long acting insulin and is available as human recombinant protamine zinc insulin product or can be found as a 100% beef insulin through few compounding pharmacies. ProZinc® (Boehringer Ingleheim Vetmedica) is the commercially available human recombinant product; it comes in a concentration of 40 U/mL (U 40). 100% beef PZI is available from BCP Veterinary Pharmacy, Houston, TX and possibly other compounding pharmacies. BCP will compound 100% beef PZI in the following concentrations: 40 U/mL (U 40), 50 U/ML (U 50), and 100 U/ML (U 50). Most cats need twice daily injections with PZI (regardless of formulation). The beef based product may have a longer duration of action than the human recombinant insulin. Various studies have shown that PZI is effective in the management of feline DM (Gossens MM, et.al. 1998; Nelson RW, et.al. 2001; Norsworthy G, et.al. 2009; Nelson RW et.al. 2009).

Glargine insulin (Lantus, sanofi-aventis US, Bridgewater, NJ) is a long acting human insulin analog and comes in a concentration of 100U/mL (U 100). In humans glargine insulin is used to give a basal control of insulin – marketed as an every 24-hour injection – to be used with additional short acting insulin to control mealtime spikes. In humans it is released into the blood stream at a relatively constant rate (peakless insulin). In cats it has been demonstrated to have a peak activity at approximately 16 hours and its duration can be as long as 24 hours based upon blood glucose curves (Marshall RD et.al. 2002). In another study the duration of action of glargine insulin in 10 healthy cats using an isoglycemic clamp method of monitoring was a mean of 11.3 +/- 4.5 hours and time to peak action was 5.3 +/-3.8 hours (Gilor C, et.al. 2010). Most cats seem to still require twice daily dosing for optimal control. One study demonstrated that the duration of action of glargine insulin in cats. One study showed that glargine insulin in addition to a low carbohydrate diet led to a higher rate of diabetic remission in cats that those treated with protamine zinc and lente insulins (Marshall RD et.al. 2009).

Detemir insulin (Levemir®, Novo Nordisk Inc, Princeton, NJ) is a long acting, U 100, human insulin analog and similar to glargine is used as a basal, once daily insulin in conjunction with other short acting insulin for postprandial glycemic spikes. In people detemir insulin may have more predictable pharmacodynamics than glargine insulin and may help prevent undesired weight gain as a result of its albumin binding increasing the availability of the insulin to organs like the liver. In one study using 10 healthy cats the pharmacodynamics of insulin detemir and glargine were compared (isoglycemic clamp method). Mean onset of action between the two insulins differed significantly (1.8 +/-0.8 for detemir and 1.3 +/-0.5 for glargine). End of action (13.5 +/-3.5 hours for detemir and 11.3 +/- 4.5 hours for glargine) and time to peak action (6.9 +/- 3.1 hours for detemir and 5.3 +/- 3.8 hours for glargine) were not significantly different Gilor C et.al. 2010). In a 2009 ACVIM proceedings detemir insulin was shown to be effective in producing remission in 12/18 diabetic cats concurrently treated with a low carbohydrate wet food diet using a protocol (described below) for intensive blood glucose control with home monitoring. Median maximum dose of detemir insulin was 1.75 U/cat twice daily (Roomp K et.al. 2009).

Recommended initial insulin doses range from 0.25-0.5 U/kg or 1 U/cat for most insulins (PZI, Glargine, NPH). It is advised to start at the lower end of the dosing range and adjust upward as needed to avoid hypoglycemic episodes.

Client education plays an essential role in the successful management of feline DM. Clients need to know of the time (possibly life-long therapy, monitoring) and monetary (supplies, rechecks) commitment needed especially in initial stages of management. Insulin injections and insulin handling should be carefully explained, demonstrated in hospital and practiced by the owner. Cat owners may be directed to websites that provide information on DM in cats and that explain how to perform injections (www.felinediabetes.com, www.sugarpet.net, www.db.com, www.partnersah.vet.cornell.edu/pet/fhc/diabetes, etc). Veterinarians may provide clients with handouts explaining the disease, dietary management, insulin handling and administration. Clients also need to know how to monitor their diabetic cat at home, especially for hypoglycemia and what to do if they suspect an animal is experiencing hypoglycemia including what to do in case of an emergency. It is also important to explain the differences in managing human and feline diabetes as owners may have knowledge of human diabetes management.

Monitoring

Monitoring should be performed ideally every 2-3 weeks after an insulin adjustment. It can take several weeks on an insulin and diet regimen for glycemic control to improve. Once adequate glycemic control has been achieved rechecks may decrease to every 6-8 weeks then every 4-6 months in the long-term stable diabetic patient. Monitoring should incorporate owner reports of persistent or resolved clinical signs, physical examination (hydration status, weight, evidence of neuropathy), urinalysis, fructosamine or blood glucose curves. Owners may monitor urine glucose/ketones at home. Urine glucose monitoring is advised only to evaluate for possible hypoglycemia/diabetic remission (persistently negative urine glucose). Increasing insulin based only on persistently elevated urine glucose is not advised. Serial blood glucose curves are essential in identifying periods of hypoglycemia that may be followed by hyperglycemia (Somogyi phenomena). Blood glucose curves aid in determining the duration and efficacy of insulin action as well as the nadir. Curves generated in hospital may be of questionable value, however, because of stress hyperglycemia and in-hospital anorexia. Many owners may be receptive to performing serial blood glucose curves at home, which may provide better information with less interference due to stress. Portable glucometers should be compatible for use in cats and owners should know how to perform calibrations for quality control. The AlphaTRAK (Abbott Animal Health, Maidenhead, England) was specifically developed for use in cats and dogs. It is costly, but may be superior to human glucometers (Zini E, et.al. 2009). Human glucometers may read lower when compared to automate serum analyzers. If blood glucose curves cannot be generated, fructosamine can be used as a tool for monitoring. Fructosamine should be interpreted in light of owner reports of clinical signs and physical examination, paying close attention to weight loss or gain and signs of diabetic neuropathy. Fructosamine levels that are high indicate that mean blood glucose level has been elevated over the past week or weeks, but not why. It could mean that there is a somogyi phenomenon occurring or that insulin dose is insufficient. Normal fructosamine may not reflect episodes of hypoglycemia. Insulin dose is usually increased in 0.5 to 1U increments per cat. If hypoglycemia is detected insulin dose should be decreased by 25-50%. As diabetic remission is possible in cats, especially if early and tight glycemic control is instituted, an intensive monitoring protocol (www.uq.edu.au/ccah for details regarding intensive monitoring protocol) has been developed for use with glargine insulin and a low carbohydrate diet (Roomp K et.al. 2009). This requires that owners perform at home blood glucose monitoring daily in an effort to keep blood glucose between 50-200 mg/dL throughout the day. An owner may be directed to www.uq.edu.au/ccah for instructions on how to obtain blood for home monitoring. An overall remission rate was seen in 64% of cats starting this protocol. A significant difference was detected in remission rates in cats started on the intensive protocol within 6 months of diagnosis (84%) vs those started after 6 months diagnosis (35%).

Troubleshooting Insulin Resistance

In insulin-dependent diabetic cats insulin resistance is manifested as an inadequate response to an appropriate pharmacologic dose of insulin. There is no specific dose that is diagnostic for insulin resistance; however most cats can be control on insulin doses ranging from 1-3 U per dose. Cats requiring doses greater than 1.5 U/kg per dose or > 6 U per dose to achieve good control or are persistently hyperglycemic despite these doses should be evaluated for insulin resistance. However before investing monies in additional testing it is imperative to assure that there are no owner compliance issues and that a Somogyi phenomenon is not occurring.

Concurrent diseases that may be associated with insulin resistance include obesity, pancreatitis, hepatic lipidosis, cholangiohepatitis, urinary tract infection, renal failure, hyperthyroidism, hyperadrenocorticism, inflammatory bowel disease, acromegaly, and heart disease. Treatment with exogenous glucocorticoids or progestagens can also cause insulin resistance. In some cases, the severity of insulin resistance can be overcome by increasing the dose or changing insulin formulation to a more potent product; however, in other disease processes, such as acromegaly, insulin resistance is severe and cannot be overcome with even extremely large doses of insulin.

Pancreatitis, especially chronic pancreatitis, may contribute to the pathogenesis of diabetes mellitus in cats. Chronic inflammation due to pancreatitis causes insulin resistance that may impair glycemic regulation. Compounding the problem of insulin resistance is the cyclic nature of the disease causing fluctuating insulin requirements and appetite with fluctuating severity of inflammation. This leads to an increase in hyper- and hypoglycemic episodes. Diagnosis can be difficult and relies on clinical signs, physical examination findings, ultrasound, and serum pancreatic lipase immunoreactivity. Treatment is supportive and prognosis for long-term resolution is guarded.

Cats with diabetes are at risk of infection, especially urinary tract infection and many may not exhibit clinical signs. Urine culture is an important part of the diagnostic work-up for insulin resistance. Skin, oral cavity, and biliary tract are other sites that can commonly support bacterial infections. It has been hypothesized that diabetic patents have impaired humoral and cell-mediated immunity, abnormal neutrophil chemotaxis, and defects in phagocytosis.

Renal disease is common in diabetic cats. This may occur secondary to DM or may be a concurrent disorder. Cats with moderate to severe renal failure may cause insulin resistance and may also be at an increased risk for hypoglycemia due to decreased insulin clearance. Physical examination, CBC, serum biochemical profile, urinalysis, urine culture, UPC and ultrasound of the urinary tract are helpful in assessing for renal failure.

Hyperadrenocorticism (HAC) is a rather uncommon but important cause of insulin resistance in cats. HAC can be pituitary dependent (more common) or can be caused by an adrenal tumor. Approximately 80% with HAC are diabetic at the time of diagnosis. Clinical finding in cats are similar to those reported in dogs (PU/PD, polyphagia, muscle atrophy, lethargy, alopecia, pot-bellied appearance, hepatomegaly, and cutaneous fragility. Skin fragility may be so severe that tearing of the skin may occur during physical examination or routine grooming. On biochemical profile increased ALT, ALP, cholesterol, hyperglycemia, low BUN may be documented. Testing to screen for HAC in cats includes ACTH stimulation test, low-dose dexamethasone suppression using a higher dose of dexamethasone (0.1mg/kg intravenously), urine cortisol:creatinine. ACTH stimulation is neither particularly sensitive nor specific in cats but may be useful is LDDS testing is difficult to interpret. In cats with adrenal tumors, cortisol secretion may not be a major player and concentrations may be low with minimal response to ACTH. A sex hormone profile with ACTH stimulation may be considered. Abdominal ultrasound, high-dose dexamethasone suppression, and endogenous ACTH may help for differentiation. No studies have evaluated the performance of these tests for differentiation in cats. Treatment options include adrenalectomy for adrenal tumors, bilateral adrenalectomy for PDH. Some cats will respond to treatment with trilostane. Hypophysectomy or radiation therapy can be considered as well.

Acromegaly is caused by excess secretion of growth hormone (GH) from a pituitary adenoma. GH causes insulin resistance, carbohydrate intolerance, hyperglycemia, and DM. Excess GH results in increased secretion of insulin growth factor 1 (IGF-1) from the liver and peripheral tissues. IGF-1 has anabolic effects causing proliferation of bone, cartilage, and soft tissues resulting in organomegaly. Cats often have a large body size, weight gain despite poor glycemic control and enlargement of the head and extremities. Respiratory stridor can be noted caused by enlargement of the tongue and oropharyngeal tissues. These cats can tolerate very high doses of insulin. Cats can have central neurologic signs resulting from the enlarging pituitary mass. Some cats with growth-hormone secreting tumors may not have the typical phenotype of an acromegalic cat and cats with very high insulin requirements should be evaluated for acromegaly. Serum IGF-1 concentrations can be used as a screening tool. Assays for GH have been validated in the cat. Contrast CT or MRI can confirm acromegaly. Radiation therapy can be effective for treatment. It has been reported to improve neurologic signs and decrease insulin requirement in cats (Sellon RK et.al. 2009; Dunning MD et.al.2009). Transphenoidal hypophysectomy was used successfully to treat one acromegalic cat (Meij BP, et.al. 2010).

References

1. Panciera DL, et al. JAVMA 1990;197: 1504.

2. Rand JS, et.al. Australian Veterinary Journal 1997;75:402-405.

3. Feldhahan JR, et.al. J Feline Med Surg 1999;1:107-115.

4. Appleton DJ, et.al. J Feline Med Surg 2001;3:211-228.

5. Link KRJ, et.al. JVIM 1996;10:185.

6. Feldman EC, Nelson RW. Canine and Feline Endocrinology and Reproduction. St. Louis: Saunders; 2004.

7. Roomp K, et.al. Rand, JS. J Feline Med Surg 2009;11:668-682.

8. Mayer-Roenne B, et.al. J Feline Med Surg 2007;9: 124-132.

9. Nelson RW, et.al. Proceedings of the 12th ACVIM Forum, San Francisco, CA, USA, p. 996.

10. Bennett N, et.al. J Feline Med Surg 2006;8:73-84.

11. Marshall RD, et.al. JVIM 2002;16:358.

12. Gilor C, et.al. JVIM 2010;24:870-874.

13. Marshall RD, et.al. J Feline Med Surg 2009;11:683-691.

14. Roomp K, et.al. (abstract). JVIM;23: 697.

15. Zini E, et.al. Schweiz Arch Tierheilkd 2009;151:448-451.

16. Roomp K, et.al. J Feline Med Surg 2009;11:668-682.

17. Sellon RK, et.al. J Vet Intern Med 2009;23:1038-44.

18. Dunning MD, et.al. J Vet Intern Med 2009;23:243-249.

19. Meij BP, et.al. J Feline Med Surg 2010; 12:406-410.

20. Rois L, et.al. Compendium 2008;30:E1-E7.

21. Rand JS J Feline Med Surg 1999;1:143-153.

22. Rois L, et.al. Compendium 2008. 626-640.

23. Rand JS J Feline Med Surg 1999;2:3-17.

24. Scott-Moncrieff JC. Vet Clin Small Anim 2010;40:241-257.

25. Rucinsky R, et.al. J Am Anim Hosp Assoc. 2010;46:215-224.