Abstract

A 21-year-old male orangutan was diagnosed as having Type II (non-insulin dependent) diabetes mellitus in 1996. In the ensuing 8 years, his case has been managed by a combination of hypoglycaemic medication, daily insulin injections and dietary manipulation. This paper will outline the approach to monitoring and treatment, the introduction of daily insulin injections and blood glucose testing, the use of oral hypoglycaemic medications, and the decision process for making dose changes.

Diagnosis of Type II Diabetes Mellitus and Initial Dietary Manipulation

“Hsing Hsing,” a male Sumatran orangutan, has been in the Perth Zoo collection since 1983, when he arrived at the age of 9 years. Prior to 1996, there had been no clinical or physical evidence of diabetes mellitus. The disease was first suspected when hyperglycemia (18 mmol/L) and marked (4+) glycosuria were detected from samples taken under anaesthesia during routine physical examination. Subsequent regular testing of voided urine via dipstick (Multistix urinalysis reagent strips, Bayer Australia Ltd., Pymble, NSW) revealed that the glycosuria was persistent.

In 1996, Hsing was 21 years old, and of good general health. Prior to arrival at Perth Zoo, he had been on a highly inappropriate and unregulated diet, which included regular access to confectionery and sugary drinks. Although he did not appear overweight when he arrived, his parents, being on a similar diet, were certainly obese. It was later discovered that Hsing’s father was also diagnosed as having diabetes, although specific details of his clinical progress are not available.

Hsing’s diet in 1996 consisted of a mixture of fruits and vegetables, eggs, bread and primate pellets, with regular access to browse. Foods used for enrichment included nuts, sultanas, peanut butter and jam. In response to the suspicion of diabetes, Hsing’s diet was altered immediately to decrease its total caloric and sugar content. This was achieved by increasing browse and vegetables, restricting fruit (particularly tropical varieties), and eliminating behavioural enrichment foodstuffs except for diet jams.

After 2 weeks on the revised diet, no changes were seen in Hsing’s voided urine glucose levels. Further information on Hsing’s clinical condition was required in order to identify more specific treatment options. The following tests were performed on blood collected under anaesthesia in November 1996.

Fasting C-peptide levels: within normal human limits (0.44 nmol/L, reference range 0.20–0.90). This suggested that Hsing was producing insulin precursors at the normal rate.

Assays for antibodies to pancreatic β-islet cells and glutamic acid decarboxylase (GAD): negative. A negative result indicates an absence of pancreatic autoimmunity, which in turn suggests normal pancreatic function.

Attempts to separate a glycosylated hemoglobin fraction (HbA_1c), which becomes elevated in the presence of ongoing (>4 weeks) hyperglycemia, were unsuccessful. The findings were considered by diabetologists to be consistent with a slowly evolving, type 2 (non-insulin dependent) diabetes mellitus.

Initial Treatment with Oral Hypoglycemics

Hsing was not showing any overt clinical signs of diabetes at the time of deciding to start him on oral hypoglycaemic medication. His appetite was the only evidence of a potential subclinical condition, being only fair to moderate, which was reflected by his average to lean body condition, and slightly low body weight of 81 kg; however, he was not showing evidence of polydipsia or polyuria

Given the absence of clinical manifestations of diabetes, diabetologists felt that there was scope to manage Hsing’s condition without insulin therapy. Treatment was begun with metformin (Diaformin 500 mg tablets, Alphapharm Pty. Ltd., Glebe, NSW) in early 1997. Metformin acts to enhance the effect of insulin. It promotes increased glucose uptake in the presence of insulin but does not elicit hypoglycemia.¹ It was considered that metformin would constitute a means of safely initiating a decrease in blood glucose levels without the risk of hypoglycemia. Hsing readily accepted tablets crushed up and mixed with diet cordial.

In selecting a medication regime there was concern about the limited alternatives for monitoring Hsing’s response to medication. At this point, he was not trained to tolerate conscious blood sampling, so the only clinical pathology parameter available for quantifying his condition was dipstick urinalysis of voided urine.

Prior to treatment, Hsing’s urine dipsticks showed a consistent, marked glycosuria (3–4+), occasional traces of protein, and occasional ketones. The medication aim was to reduce the glycosuria to 2+ and eliminate the episodes of ketonuria. The limitations of monitoring via dipstick urinalysis were recognised, and keepers began working on conditioning Hsing for injections and blood sampling, using the traditional principles of operant conditioning.

The metformin dose was gradually increased from 125 mg SID to 500 mg SID over 3 weeks, in order to observe for any untoward side effects of the medication, such as gastrointestinal upsets. After 1 month on metformin, Hsing was anaesthetised for reassessment. While his general health continued to be good, his body weight had declined to 76.5 kg. Hematology and serum biochemistry were normal, with the exception of continued hyperglycemia (11 mmol/L) and elevated total bilirubin (28.0 µmol/L). The hyperglycemia was less marked than previously noted. However, urine dipstick glucose was unchanged at 3–4+, although there had been no evidence of ketones reappearing in the urine.

A second oral hypoglycemic agent was added to Hsing’s medication regime in May 1997. Gliclazide (Diamicron 80 mg tablets, Servier Laboratories Aust. Pty. Ltd., Hawthorn, Vic.), a sulfonylurea hypoglycemic agent, stimulates insulin secretion from functional pancreatic β-cells, and also increases their sensitivity to glucose stimulus.¹ Sulfonylurea agents also have extrapancreatic effects on lowering blood glucose. The initial dose of gliclazide was 20 mg SID, given together with 500 mg SID of metformin. Over the ensuing month, this dose was increased to 40 mg SID due to continuing 3–4+ glycosuria.

By mid-1997, Hsing was on 60 mg of gliclazide and 500 mg metformin daily. Although his body weight had increased to 81.5 kg, his blood glucose was still very high (15.7 mmol) at anaesthesia in early 1998. Keepers were now making progress with conditioning Hsing to present his fingertip for blood sampling, and his shoulder for injections, using recognized operant conditioning techniques.^2,3 Hsing’s gradual weight gain, and the continued absence of ketonuria, indicated some improvement in his condition.

Although the hyperglycemia persisted, the decision was made to avoid insulin therapy at this point. Hsing still had no overt polyuria or polydypsia, and his daily oral medication doses were still much lower than the daily maximum doses recommended for humans (1000 mg TID of metformin and 160 mg BID of gliclazide). Furthermore, Hsing’s cooperation with presenting his shoulder for injections was still unreliable, so insulin could not be considered immediately. Veterinarians elected to manage his condition by gradually increasing his oral medication dosages, provided that no side effects were seen.

Clinical Manifestations of Diabetes Emerge

By July 1999, keepers had Hsing trained to tolerate finger pricking for blood glucose monitoring (Esprit Glucometer, Bayer Australia Ltd, Pymble NSW). Hsing’s blood glucose levels were usually between 20–30 mmol/L, and never below 10 mmol/L. Hsing’s HbA_1c, measured from finger prick blood samples, increased from 12.5% in August 1999 to 13.6% in November 1999. Although normal HbA_1c levels have not been determined in orangutans, these levels are considered overtly diabetic according to the analyzer specifications for humans (DCA 2000 Hemoglobin A_1c Reagent Kit and Analyzer, Bayer Corporation).

At this time, Hsing began to exhibit symptoms of sweating, restlessness and erratic behaviour. These symptoms were interpreted as clinical manifestations of persistent, chronic hyperglycemia, and reflected his elevated blood glucose readings and increasing HbA_1c. Hsing’s gliclazide dose was increased to the daily maximum of 160 mg BID. As these symptoms persisted, gradual increases in the doses of metformin were also instigated, reaching 1000 mg BID in late 1999. In spite of these relatively high medication doses, Hsing’s blood glucose readings persisted at 15–30 mmol/L.

In the light of the emergence of clinical signs, and the poor response to high doses of oral hypoglycemic medication, it was decided that treatment with insulin was now unavoidable. Hsing was anaesthetised to review his general health, and to examine him for side effects of chronic hyperglycemia.

Hsing’s fasting insulin levels were 6 mU/L (normal human range 3–26 mU/L), indicating a normal level of production of insulin. C-peptide levels continued normal, and there was once again no evidence of anti-islet cell or GAD antibodies. Examination by a veterinary ophthalmologist did not detect any evidence of diabetes-related retinopathy. Serum microalbumen assay did not demonstrate any evidence of renal damage associated with chronic diabetes. His body weight at this time was 86 kg.

Initiation of Insulin Therapy

Insulin medication commenced in March 2000. Due to the difficulties in coordinating insulin treatment with Hsing’s daily routine, an intermediate acting human insulin, with a duration of effect of 24 hours, was selected (Protaphane, isophane human insulin injection, Novo Nordisk Pharmaceuticals Pty. Ltd. Australia, North Rocks, NSW). The onset of effect of the isophane insulin was 1.5 hours, with a maximum effect at 4–12 hours.¹ The insulin was administered to Hsing immediately before his evening meal.

Unfortunately, Hsing became refractory to blood sampling at this time, forcing a reliance on dipstick urinalysis and behavioural cues to try and detect a response to treatment. The starting dose of 4 U SID was increased in 4 U increments on a weekly basis. At the onset of insulin therapy, the gliclazide dose was halved to 160 mg SID rather than 160 mg BID, since the combination of a high gliclazide dose with insulin therapy increased the risk of hypoglycemia.¹

By May 2000, the insulin dose was 20 U SID. At this time, urine glucose levels registered 2–3+ consistently rather than 3–4+ as had been seen on previous dose regimes. Gliclazide and metformin doses continued at 160 mg SID and 1000 mg BID respectively.

The clinical response to the insulin treatment was significant. Hsing became less irritable, and his activity and alertness increased. Polyuria and polydypsia had resolved completely by May 2000, and there was a significant improvement in his appetite. By June his weight was 90 kg, and by early September had reached his target weight of 95 kg.

Incremental increases of 4 U in the insulin doses continued until a dose of 24 U was reached in May 2000. By mid-June 2000, keepers had regained compliance for blood glucose testing. Shortly after this, an excessively low evening blood glucose level of 2.7 mmol/L was registered. It was thought that the combination of the morning glipizide dose with the insulin therapy was resulting in the potential for evening hypoglycemia, so the daily gliclazide dose was halved to 80 mg.

Now that a therapeutic dose of insulin had been reached, the aim of treatment was to maintain blood glucose levels at 5–10 mmol/L. Hsing’s blood glucose levels tended to be lower in the evening (6–7 mmol/L) than in the morning (13–14 mmol/L). Since the evening glucose levels were still too high, an insulin dose increase was warranted, but this carried some risk of eliciting morning hypoglycemia. Therefore, from June 2000, the gliclazide dose was gradually reduced, and the insulin dose increased, until August 2000, when gliclazide therapy was withdrawn completely. The insulin dose had been increased to 34 U by this time. Since gliclazide can cause side effects of gastrointestinal disturbance and biochemical abnormalities (including elevations in CK, ALP, AST, BUN and bilirubin),¹ this was considered to be an improvement in Hsing’s long-term medication plan.

A review of the metformin dose was conducted in November 2000. Because this drug increases the effectiveness of insulin,¹ an extra 500 mg midday dose was initiated. This brought the total daily dose of metformin to 2500 mg. It was hoped that this would reduce the requirement for further increases in the insulin dose. This did seem to have some effect, and the insulin dose was not increased again for another month. The insulin dose rose by a further four units through 2001 to 38 U per day.

Introduction of a New Oral Hypoglycemic Agent

In late 2001, another oral hypoglycemic agent was added to the medication regime. Glipizide (Minidiab 5 mg tablets, Pharmacia & UpJohn Pty. Ltd., Rydalmere, NSW) acts on β islet cells to stimulate production of endogenous insulin.¹ This was likely to elicit a response in Hsing’s case, since insulin assays and anti-islet cell antibody assays indicated that he had a functional β-islet cell population. Glipizide was initiated at 2.5 mg SID in the morning, increased to the therapeutic dose of 5 mg SID when no side effects were seen after 14 days. The use of glipizide seems to have been successful in stimulating endogenous insulin production, in that Hsing’s insulin dose has increased by only 6 U in the 2.5 years subsequent to the introduction of glipizide.

Discussion

Predisposing factors in the development of type II diabetes include age, obesity and genetics.⁵ While Hsing was never obese, he certainly had a very high sugar diet in his early years, which may have prompted the expression of his genetic predisposition for type II diabetes. In the early stages of development of type II diabetes, peripheral insulin resistance induces hyperinsulinemia, which eventually gives way to hypoinsulinemia as the pancreas fails to maintain adequate insulin secretion.⁵ The development and manifestation of Hsing’s disease was consistent with that of a “borderline” type II diabetic: hyperglycemia, weight loss, polyuria and polydypsia developed gradually over 3 years. C-peptide and insulin assays did not detect a hyperinsulinemic phase, but this may have occurred prior to diagnosis.

Although it was not possible to monitor blood glucose in the early stages of oral medication, it appears that Hsing’s diabetes did not improve dramatically until insulin therapy was initiated. As well as the decline in his blood glucose levels, his improvement was marked by significant increases in alertness, activity, appetite and body weight.

At the time of writing, Hsing’s medication regime consisted of isophane insulin at 44 U/day (given in the evening), metformin at 2500 U/day (divided into 8 am 1000 mg; 12 noon 500 mg; 4 pm 1000 mg), and glipizide 5 mg each morning. His insulin levels have not increased since January 2003, and oral hypoglycemic doses have not changed since 2001. His blood glucose is checked by keepers every third day, in the morning and evening, and the veterinary department perform a regular review of his average blood glucose levels, originally on a monthly basis but now generally every 2–3 mo. Current blood glucose levels range from 6–13 mmol/L, with evening levels generally slightly lower than morning levels. As has been the case with the long-term management of diabetes in other nonhuman primates,⁴ there is some variability in Hsing’s response to insulin, and occasional hyperglycemic spikes are seen. Often, the cause remains unidentified, although there is some correlation with stressful episodes such as hot weather.

A feature of Hsing’s long-term management has been the establishment of contingency plans for managing his diabetes in the event of changes in his routine. Veterinarians have worked with diabetologists, dietitians and diabetes nurses to formulate contingency plans for keepers and veterinary staff. Ongoing communication with these experts continues to be an important feature of the management of Hsing’s case. Contingency planning is now in place for the following events.

Hypoglycemic Episodes

In the initial stages of treatment, when blood glucose monitoring was not possible, it was feared that gliclazide medication might trigger a hypoglycemic episode. A protocol was generated, outlining the signs of hypoglycemia, first aid actions for keepers, and veterinary emergency procedures.

Non-Compliance

If Hsing refuses to eat in the morning, glipizide medication is withheld. Because metformin does not elicit hypoglycemia, this is given regardless of his appetite. If he refuses to comply with insulin injection, keepers are instructed to persist for as long as possible. Missing the occasional dose of insulin is unlikely to be life-threatening in the type II diabetic, but fortunately Hsing’s compliance has been 100% since May 2000. If Hsing refuses his evening meal subsequent to receiving his insulin injection, he is left with access to food overnight. All episodes of non-compliance are managed on a case-by-case basis in consultation with Perth Zoo veterinarians.

Elective GA Management

In the case of elective anaesthesia, metformin treatment is stopped 3 days before surgery to prevent lactic acidosis; glipizide and insulin are continued as per normal; procedures are scheduled for early morning; consideration is given to an intra-operative dose of insulin if indicated by high blood glucose readings. Metformin is reinstated 2 days post-GA.

Acknowledgments

The authors would like to sincerely thank all the diabetes specialists who have donated their time and resources to Hsing’s case, particularly Mrs Beryl Marsh, Dr David Hurley and the other staff of the Diabetes Clinic at Royal Perth Hospital; Mr Stuart Moody (Bayer Australia); and the staff of the primate section and veterinary department at Perth Zoo.

Literature Cited

1.  Casswell, A. (ed). 1999. 1999 MIMS Annual (Australian Edition). MediMedia Australia Pty. Ltd., St Leonards, NSW.

2.  Laule, G.E., Thurston, R.H., Alford, P.L. and Bloomsmith, M.A. 1996. Training to reliably obtain blood and urine samples from a diabetic chimpanzee (Pan troglodytes). Zoo Biology 15: 587–591.

3.  Stringfield, CE and McNary JK. 1998. Operant conditioning of diabetic primates to accept insulin injections. Proc AAZV 1998 396–397.

4.  Sutherland-Smith, M. 2002. Management of diabetic primates at the San Diego Zoo. Proc AAZV 2002: 442–443.

5.  Walzer, C. 1999. Diabetes in primates. In: Zoo and Wild Animal Medicine, Current Therapy (4th ed), ME Fowler and RE Miller (eds). WB Saunders, Philadelphia. 397–400.