Pet Health: BG Monitoring   Pet Supplies



  Cat Health and Supplies
A Rational Approach to Feline Blood Glucose Curves
G.D. Norsworthy, DVM, Dipl. ABVP
Acres North Animal Hospital
San Antonio, Texas

The glucose curve is the most effective way to monitor insulin therapy in diabetic cats. But curves are expensive, and many consider euthanasia because of the high cost. In such cases, here are some practical, less expensive alternatives.

From the November 1995 issue of Veterinary Medicine (a peer-reviewed journal)

Diabetes Mellitus in cats is one of the most frustrating diseases faced by veterinarians. Diabetic cats have certain peculiarities not seen in dogs, including the marked effect of stress on glucose concentrations. In addition, a cat's response to insulin is much less predictable than a dog's. We know, in general, what to expect from the different types of insulin. But the same type of insulin may be absorbed and metabolized differently from one cat to the next. Assumptions about peak times and duration of action in diabetic cats are often inaccurate. The only way to know how any given insulin works in an individual cat is to perform a glucose curve.

The standard glucose curve and its alternative

A glucose curve is a series of blood glucose determinations made after a dose of insulin is given. Typically, blood samples are taken every 1 1/2 to 2 hours until the effects of the insulin injection can be determined. For ease of understanding, they are often plotted on a graph (Figures 1 & 2).

I use the term "mini-glucose curve" to describe blood glucose determinations made just before an insulin injection is given and at the previously determined peak time This two-point curve should identify the highest and lowest, or peak and trough, blood glucose concentrations. As described below, in certain situations, the mini-glucose curve is a useful substitute for the full glucose curve.

Fig 1. This is an example of an ideal glucose. It has its highest point below 300 mg/dl and its lowest point, the nadir, above 100 mg/deciliter. The range is from 105 to 297 mg/dl, and the range midpoint is 201 mg/deciliter. The peak time is six hours. These criteria mean that this insulin type and dose should produce good regulation when given every 12 hours.


Fig 2. This glucose curve is not suitable. The highest point is 380 mg/deciliter. The nadir is 160 mg/deciliter. The range is 160 to 380 mg/dl, and the range midpoint is 270 mg/deciliter. The peak time is at 4 p.m., or eight hours after the insulin injection. Three blood glucose determinations were made after the nadir, but the concentrations at 4 p.m. and 4:30 p.m. were so similar that it was decided that two subsequent determinations should be made after those two values to verify that the slope was rising. Based on these results, the type of insulin is correct because the peak time occurs eight hours after the insulin injection. The insulin dose be increased because the midpoint of the range is 270 mg/dl and the nadir is not below 100 mg/deciliter.

Indications for a blood glucose determination

There are three indications for determining blood glucose concentrations:

  1. To establish initial insulin protocol at the time of diagnosis
  2. To monitor the degree of regulation and
  3. To rule in or out rebound hyperglycemia.

Initial regulation

The first indication for a glucose curve is to establish the insulin dose dosing interval, and insulin type during the initial regulation process. The conventional way to regulate a diabetic cat is to choose an insulin type dose, and dosing interval; adjust insulin for at least three days; and perform the glucose curve. If the results indicate that any of these variables is inappropriate, the regimen is altered. The new regimen is then used for at least three more days and the glucose curve is repeated. It is not unusual for three to five glucose curves to be performed before a satisfactory regimen is determined.

Because of the expense of performing multiple glucose curves, many owners are unable to afford the initial regulation process. In lieu of that many diabetic cats are euthanatized. To avoid this unfortunate situation, I have developed an alternative approach in cats that have received the initial conservative dose of insulin and are still consistently showing signs of diabetes, including polyphagia and polyuria. The initial dose is presumed to be inadequate if the blood glucose concentration just before insulin administration is greater than 350 rng/deciliter. In such cases, the serial blood glucose determination is discontinued. The insulin concentration is increased by 10 to 20% and administered for at least three more days, when the cat is returned for another full glucose curve. This process is repeated until the initial blood glucose concentration is below 350 mg/ deciliter. At this time, a full glucose curve is performed.

The goal for regulating a diabetic cat is to keep the blood glucose concentration between 100 and 300 mg/dl throughout the day. If the blood glucose concentration at the time of diagnosis is less than 400 mg/dI, insulin is begun at 1/4 unit/lb given subcutaneously twice daily. An intermediate or long-acting insulin is given. If the blood glucose is more than 400 mg/dI, 1/2 unit/lb is given twice daily.

Monitoring regulation

The second indication for a glucose curve is to monitor a diabetic patient's level of regulation. A glucose curve should be performed at regular intervals as determined by the attending veterinarian (usually every three to six months). A full glucose curve should be performed whenever signs of diabetes or dysregulation occur (erratic fluctuations in clinical signs or behavior from one day to the next). However, if the cat is presented for a routine follow-up and appears well regulated or if there are serious financial constraints that prevent a full glucose curve, I often perform a mini-glucose curve. I recognize that there is some variation in glucose curves from one day to the next, which limits the effectiveness of the mini-glucose curve. But financial constraints often become such an important factor with many owners of diabetic cats that a less expensive alternative to a full glucose curve is often a necessary option.

Documenting rebound hyperglycemia

The third indication for a glucose curve is when rebound hyperglycemia is a consideration. This phenomenon occurs when blood glucose declines to life-threatening concentrations, generally below 65 mg/ deciliter. In response to this crisis, gluconeogenesis and glycogenolysis occur, resulting in glucose formation. The blood glucose concentration rises quickly, but the cat is unable to produce the appropriate amount of insulin because it is diabetic. Within a few hours, the blood glucose concentration may exceed 600 mg/dl and produce the classic signs of diabetes that occur in response to hyperglycemia.1 The best way to document this occurrence is with a full glucose curve. In this situation, the use of the mini-glucose curve is inappropriate.

Contraindications for blood glucose determinations

The primary contraindication for performing a glucose curve is a very fractious cat. Because of the stress response that occurs in cats, blood glucose determinations in such cats are very inaccurate. Sometimes, this situation can be managed by hospitalizing the cat for a few days to acclimate it to the hospital environment. During this time, the cat should be isolated from the sights and sounds of dogs and other cats. The hospital staff should make concerted efforts to befriend the cat. It should be petted, stroked, and talked to in an effort to ease its fear.

Another approach that may work is to do the glucose curve on an outpatient basis. This is successful only if the owner lives nearby and the cat does not become agitated by riding in the automobile. In such cases, the cat is presented for blood collection and taken home. This process is repeated until the glucose curve is completed. The reception staff should be told that this patient should not be kept waiting in the reception area, especially if dogs or other cats are there. Upon arrival, the owner and cat should be sent directly to a quiet, closed examination room, and the blood sample should be taken quickly with as little stress and restraint as possible.

The other contraindication for a glucose curve is the concurrent administration of any drug that would affect glucose concentrations. These drugs include dextrose, glucocorticoids, and progesterones. There is little reason for doing a glucose curve when a cat is receiving intravenous dextrose. Glucocorticoids and progesterones are contraindicated in diabetic cats.

The glucose curve is the beat tool we have to understand how a diabetic cat reacts to any given type and dose of insulin. However, it is not infallible. Great emphasis should also be placed on the cat's total clinical picture. If the cat is maintaining proper weight or gaining weight and is not polydipsic, polyuric, or polyphagic, its state of regulation should be considered adequate regardless of the results of the glucose curve.

The steps in the procedure

The cat should be fed immediately before leaving home. Upon arrival at the hospital, the owner should be escorted directly to an examination room, or the cat immediately admitted to minimize waiting. The first blood sample should be taken without delay. It is important that this step not be skipped because, generally, this sample will yield the highest concentration on the glucose curve - a value you need to know and record. This approach differs from the standard recommendation of fasting the cat until after the first blood sample is taken and the insulin is administered. I prefer this approach because of the difficulty I have getting the cat to eat the proper food in the amount that it would at home. If the owner cannot bring the cat to the hospital immediately after feeding it (e.g. because it eats at 6 a.m. and your hospital does not open until 8 a.m.), the conventional approach should be taken.

I prefer using the cephalic vein when collecting blood for a glucose curve. Use of a small tourniquet, alcohol to wet the hair, a 25-ga. needle, and a heparinized syringe will permit repeated blood collections from most cats, I take my first sarnple from the right cephalic vein just above the carpus. Subsequent samples are taken from alternating legs and slightly proximal to the previous one. By using this rotational pattern, 10 samples or more can be taken in one day from most cats. Other clinicians prefer the jugular or femoral vein. I find the jugular vein harder to see without clipping hair, which most clients find objectionable. The femoral vein is much smaller than the cephalic and tends to form hematomas very easily. However, in some cats it is necessary to use a combination of all of these.

Some clinicians prefer to place a jugular catheter so that multiple blood samples can be taken without repeated venipuncture. The advantages of this approach are that blood collection is very easy and inflicts minimal or no stress on the patient. The disadvantages are the cost and technical difficulty of placing this type of catheter.

Treating a diabetic cat for several years may require that several hundred blood samples be taken. Because of the small size and limited number of accessible veins in cats, it is important that you minimize trauma to the veins. Generally, I do not place a cephalic catheter in a nonketoacidotic diabetic cat because an intravenous catheter will render the vein unusable for blood collection for several days to weeks. To minimize hematoma formation, I cover the venipuncture sites with tape after each sample. Another disadvantage of using the jugular or femoral vein is the inconvenience of applying appropriate pressure; these veins should receive direct pressure for five minutes or more to avoid hematomas.

Next, insulin is given. It should be the same dose and type as has been used for the last three days. It is highly preferable to have the owner bring his or her bottle of insulin to prevent adding another possible variable.

The blood glucose is then measured every 1 1/2 to 2 hours until the nadir (the lowest point on the glucose curve) is reached. After that, one or two blood glucose determinations are made to be sure that the blood glucose is on the increase. The second sample is taken because of possible variation in blood glucose determinations, regardless of the method used, as discussed below.

The ideal way to perform a glucose curve is to continue taking blood samples until it is time for the next insulin injection. This is feasible in hospitals staffed at least 12 hours a day, if insulin administration coincides with the hospital schedule. But most private veterinary hospitals, including mine, are not open 12 hours a day. So, because the key determinants have been made once the nadir is passed and because glucometers can produce variable readings, I prefer to make two blood glucose determinations, at 30-minute intervals, past the nadir.

Laboratory methods for making glucose readings

Several means for making blood glucose determinations are available. You can submit the samples to a reference laboratory. However, this is the most expensive way, and it will not allow you to know when adequate samples have been taken.

Blood glucose determinations can be made using dry chemistry machines, which are found in many veterinary hospitals. This method is much cheaper than using a reference laboratory, and it will permit readings to be made as the glucose curve is in progress. Some of these machines perform best using serum instead of whole blood or plasma. If so, larger blood samples are needed, and more time is required for clotting and separating the serum from the clot.

A third option is a hand-held reflectance meter. This instrument is designed for use by human diabetics. Its retail price is about $100, but it is often heavily discounted. Reflectance meters have been shown to be accurate for veterinary use. They require only two or three drops of whole blood and about 60 seconds in operation time. They are less expensive to operate than a dry chemistry machine, both in labor cost and cost of the operating supplies. However, newer models, calibrated for capillary blood are not as accurate when venous blood is used. These models should be avoided since we use venous samples.

Regardless of the method used, glucose determination by laboratory methods is not an exact science. One study found that glucose concentrations above 180 mg/dl could vary as much as + or - 90 points from one machine sampling to another. Reflecters can be misused so there is a greater degree of inaccuracy interposed. There are several ways to increase the accuracy of reflectance meters. (These are listed in Table 1) This frightening degree of variability is something that we have to recognize and consider when interpreting the glucose curve. Glucose curves with values that do not seem to be in context with the others in the curve may represent the variation that occurs in blood glucose testing. Because of this variation, we should not try to regulate a diabetic cat too closely.


Ways to Increase the Accuracy of Reflectance Meters*
  • Use test strips properly.
  • Store at room temperature.
  • Do not remove the strips from the bottle until ready for use.
  • Keep the lid on the bottle lightly.
  • Do not use strips if they are outdated.
  • Do not use generic strips.
  • Use only the right code number on the glucometer.
  • Use only the strips that are made for the reflectance meter chosen.

Keep the glucometer in good condition.

  • When the light transmission window gets dirty, clean it with water.
  • Store and use the glucometer at room temperature.

Recognize disease influences.

  • Anemia causes abnormally low or high readings.
  • Dehydration causes abnormally low readings.
  • Elevated bilirubin causes abnormally low readings.

*Adapted from Hoyson, P.M.: Diabetes 2000: Oral Medications. RN 58(5):34-39; 1995.



Interpreting the Glucose Curve*

To determine the dosing interval or the type of insulin given, evaluate the peak time.

If the peak time is:

< 5 hours: change to t.i.d. (3/day) administration or administer a longer-acting Insulin.
5-8 hours: continue b.i.d. (2/day) administration.
> 8 hours: change to s.i.d. (1/day) administration.

To determine the dose of insulin or if dysregulation is a problem, evaluate the nadir and range midpoint.

If the nadir is < 100 mg/dl, decrease the insulin dose.

The ideal range midpoint is 200 mg/dI, and 150-250 mg/dl is acceptable. If the range midpoint is:

  1. < 150 mg/dl: reduce the dose of Insulin.**
  2. 150-250 mg/dl: don't change the dose.
  3. > 250 mg/dI: several possibilities exist:
    • The cat may have been stressed when the blood glucose curve was performed.
    • The Insulin dose may be too low.
    • There may be an insulin problem (inactive insulin, insulin not mixed property, poor injection technique, poor absorption).
    • Insulin resistance may be occurring due to concurrent disease (systemic Illness, hyperthyroidism, hyperadrenocorticism, acromegaly) or insulin antibodies.

*Based on twice-daily dosing of insulin.

**The nadir also should be below 100 mg/deciliter.

Interpreting the curve

Four values are important in interpreting the glucose curve (Table 2). They are as follows:

  1. The highest blood glucose concentration, which is usually the one taken just before insulin is given. In some cases, the morning meal and the stress of riding in the car will cause the blood glucose concentration to rise slightly on the second sample, but it should decline from that point.
  2. The nadir, which is the lowest blood glucose concentration.
  3. The peak time, which is the time from the insulin injection until the nadir. It is expressed in hours.
  4. The midpoint, which falls halfway between the highest and lowest blood glucose concentrations in the range.

Once these values are determined, they can be used to determine the dosing interval, the type of insulin, and the dose. The guidelines below are based on the assumption that insulin is being given twice daily.

To determine the dosing interval and the type of insulin, follow these steps:

  • If the peak time is less than five hours, a longer-acting insulin is given, or the current insulin is given more frequently, generally three times a day.
  • If the peak time is five to eight hours, the current insulin is continued on a twice-daily schedule.
  • If the peak time is greater than eight hours, the current insulin is continued, but is given once daily.

To determine the dose of insulin or if dysregulation problems exist, use these guidelines:

  • If the nadir is less than 100 mg/dI, the dose should be decreased. You should assume that the nadir in the hospital is somewhat higher than what it will be at home because of stress. If this assumption is incorrect, you will err so that the blood glucose concentrations are too high. This is always the better way to err. Hypoglycemia can be fatal; mild to moderate hyperglycemia is not.
  • The ideal midpoint of the range is 200 mg/deciliter. However, the midpoint is satisfactory if it is from 150 to 250 mg/deciliter. If the midpoint is less than 150 mg/dI, the dose of insulin is too high and should be reduced. The nadir also should be below 100 mg/dl when the dose of insulin is too high.
  • If the midpoint is greater than 250 mg/dI, several possibilities exist: 1) The cat may have been stressed when the glucose curve was performed. Recount the cat's attitude during the blood sampling process and the hospital stay. 2) The dose of insulin may be too low; if so, increasing it will solve the problem. 3) If increasing the dose does not bring the midpoint into the desired range, insulin problems should be considered. These include inactive insulin, insulin that is not mixed properly, poor injection technique, and poor absorption. 4) If this does not seem feasible, consider concurrent infectious disease. Many infectious diseases can cause temporary insulin resistance. A two-week course of a broad-spectrum antibiotic should be considered if your physical examination and other laboratory tests do not indicate a specific problem. 5) If these approaches are unsuccessful, you should consider insulin resistance due to hyperthyroidism, hyperadrenocorticism, acromegaly, or insulin-antibody formation.


Regulating a diabetic cat can be difficult, and the glucose curve is the foundation for understanding how any given insulin works in an individual cat. It is one of the first steps in achieving regulation. It is used to determine proper insulin type and dose and proper administration interval. The glucose curve is also used as a monitoring tool for regulated cats and as a diagnostic tool for cats in dysregulation. Four values are used to interpret the glucose curve: the highest concentration, the lowest concentration (nadir), the midpoint of the range, and the peak time. Generally, these values can be determined in six to eight hours. Taking the time to properly perform a glucose curve can save a great deal of time and money for the client.


1. McMillan, F.D.; Feldman, E.C.: Rebound Hyperglycemia Following Overdosing of Cats with Diabetes Mellitus. JAVMA 188: 1426-1430:1986.

2. Joseph, R.J. et al: Evaluation of Two Reagent Strips and Three Reflectance Meters for Rapid Determination of Blood Glucose Concentrations. J. Vet. Intern. Med. 1: 170-175; 1987


Quick Navigation


Professional pet care and quality pet supplies are vital for pet health. Always consult your veterinarian about your cat's health and your choice of supplies. This site administered by a *human* physician,
and is not designed to take the place of regular veterinary care.
All content © 2004. Web Design by limestone|media