Frequency of Reevaluations

After diagnosis of diabetes mellitus, the owner should be informed that it usually takes 2 to 3 months until stable glycemic control is achieved (see Chapters 44 and 48) and that lifelong regular reevaluations in the hospital are required. Initially, frequent reevaluations should be scheduled. With time, intervals between rechecks can be extended. In our hospital, reevaluations are suggested at 1 week, 3 weeks, 8 weeks, and 12 weeks after diagnosis and then at intervals of approximately every 4 months. More frequent reevaluations may be needed in cats in which it is difficult to regulate diabetes. In animals (usually cats) in which diabetic remission has been achieved, the owners should be informed about the possibility of a relapse. They should be advised to watch their pets with regard to recurrence of polyuria and polydipsia, polyphagia, and weight loss.

Serum Fructosamine

Fructosamine measurements are widely used as an indicator of glycemic control in diabetic dogs and cats. Measurement is done in serum using commercially available test kits adapted to autoanalysis. Storage at 25° C for 5 days does not cause a significant change in fructosamine concentrations, allowing shipping of serum samples. Because fructosamine is the product of an irreversible reaction between glucose and the amino groups of plasma proteins, it is assumed that its concentration reflects the mean blood glucose concentration of the preceding 1 to 2 weeks. Serum fructosamine concentration is not affected by short-term increases of blood glucose; therefore it is also a valuable tool to differentiate between diabetes and stress hyperglycemia in cats. If healthy cats are infused with glucose to maintain a marked hyperglycemia (30 mmol/L [540 mg/dl]), fructosamine concentrations exceed the upper limit of the reference range after 3 to 5 days; if an only moderate hyperglycemia (17 mmol/L [300 mg/dl]) is induced, fructosamine concentrations mostly fluctuate just below the upper limit of the reference range (Link and Rand, 2008).

Reference ranges for serum fructosamine concentration differ slightly among laboratories but usually are approximately 200 to 360 µmol/L. In most newly diagnosed diabetic dogs and cats, fructosamine levels are greater than 400 µmol/L, but they may reach 1500 µmol/L. Normal fructosamine levels may be seen in animals with a very short duration of diabetes mellitus. Fructosamine may also be normal in animals with certain concurrent diseases. Hypoproteinemia, hyperlipidemia, and azotemia in dogs and hypoproteinemia and uncontrolled hyperthyroidism in cats can decrease the fructosamine concentration. In these cases, fructosamine measurements should not be used for the diagnosis or the long-term management of diabetes mellitus.

In all other cases, it is helpful to measure fructosamine concentrations during routine reevaluations. The parameter is independent of stress and lack of food intake, which are both major influencing factors for blood glucose measurements in the hospital. Fructosamine concentrations increase when glycemic control worsens and decrease when glycemic control improves. Differences between two consecutive measurements have to exceed approximately 50 µmol/L to reflect a clinically significant difference in glycemic control.

Normoglycemia is not a general treatment goal in veterinary medicine, and even dogs and cats with well-controlled diabetes are slightly to moderately hyperglycemic throughout the day. Consequently, fructosamine usually does not become completely normal during therapy. In contrast, the finding of a normal fructosamine level (in particular, fructosamine levels in the lower half of the reference range) should raise concern for prolonged periods of hypoglycemia caused by insulin overdose. In cats in which fructosamine levels decrease into the normal range during therapy, the possibility of diabetic remission has to be considered. Fructosamine levels of 360 to 450 µmol/L usually suggest good control; levels of 450 and 550 µmol/L, moderate control; and levels greater than 600 µmol/L, poor metabolic control. In the last situation, fructosamine measurement does not help to identify the underlying problem. All reasons for poor regulation have to be considered, including insulin underdosage, too short duration of insulin effect, diseases causing insulin resistance, and the Somogyi phenomenon.

Over the years, we and others have sometimes seen discrepancies between fructosamine on the one hand and clinical signs and blood glucose concentration on the other hand. Although high fructosamine levels suggested poor control, the lack of clinical signs and glucose level within the desired range pointed to good control. The reason for the discrepancy remains ambiguous in most instances, but it is possible that there are individual differences with regard to the extent of protein glycation. In those cases, the assessment based on clinical signs should be given priority over assessment of fructosamine.

Diabetic ketoacidosis, dehydration, acidosis, and other unidentified factors may influence fructosamine concentrations. If a diabetic patient is hospitalized for any reason, fructosamine levels measured at the time of admission may differ considerably from the fructosamine concentration measured a few days later after successful therapy. It is advisable to repeat the measurement at the time of discharge and to use this level as a reference point for future measurements.

Fructosamine is a valuable adjunct parameter to monitor glycemic control. However, it should always be interpreted in conjunction with clinical signs and body weight.

Glycated Hemoglobin

Glycated hemoglobin has been measured in human medicine since the 1970s and is regarded as one of the cornerstones of assessment of glycemic control. Comparable to fructosamine, it is formed by nonenzymatic, irreversible attachment of glucose to an amino group of the globin part of the hemoglobin molecule. Although glycation kinetics are relatively complicated, it is commonly assumed that glycated hemoglobin reflects the average blood glucose concentration over the life span of the erythrocytes, which is 100 to 120 days in dogs and 70 days in cats. Glycated hemoglobin is rarely used as a long-term parameter in veterinary medicine at the present time. The main explanation for infrequent use pertains to the various assay techniques and the lack of standardization in veterinary medicine. Therefore results may not be comparable among laboratories. Glycated hemoglobin should be considered only if the assay has been validated for dogs and cats and assay-specific reference ranges have been established.

Blood Glucose Measurements and Blood Glucose Curves

Single blood glucose measurements are usually insufficient to assess glycemic control. Exceptions to this rule are dogs and cats with an unremarkable history and physical examination and a serum fructosamine of 360 to 450 µmol/L. In these cases, the finding of a blood glucose concentration of 10 to 15 mmol/L (180 to 270 mg/dl) around the time of insulin injection would support the assumption of good glycemic control and may render further blood glucose measurements unnecessary. In all other cases, there is a substantial risk of misinterpretation if only one glucose measurement is performed. Evaluation of a BGC is mandatory in the initial phases of diabetic regulation and in animals with persistence of clinical signs, ongoing weight loss, and fructosamine greater than 500 µmol/L. The BGC provides guidelines for making rational adjustments in insulin therapy.

We usually recommend that feeding and the insulin injection be done at home and that the animal be presented to the hospital as soon as possible thereafter. If technical difficulties are suspected, owners are asked to bring the animal to the hospital before the insulin administration and to perform the whole injection procedure under the supervision of a veterinarian or a technician. Blood samples are taken every 1 to 2 hours throughout the day until the next insulin injection whether the insulin is given in the hospital or not. To avoid multiple venipunctures, we collect capillary blood from the ear or the footpad using the same sampling technique and the same PBGM as the owners at home (see later).

BGCs enable the assessment of insulin efficacy, the glucose nadir, the time of peak insulin effect, the duration of the effect of insulin, and the degree of fluctuation in blood glucose. The most important parameters are insulin efficacy, glucose nadir, and duration of effect. Insulin efficacy is defined as the difference between the highest and the lowest glucose concentration and must be interpreted in the light of the highest blood glucose concentration and the insulin dosage. A relatively small difference (e.g., 2.8 mmol/L [50 mg/dl]) is acceptable in an animal in which the highest glucose concentration is less than 11 mmol/L (200 mg/dl); however, the same difference is unacceptable if the highest glucose level is greater than 17 mmol/L (300 mg/dl). Similarly, the same difference may indicate insulin efficacy in an animal receiving a small dosage of insulin (<0.25 to 0.5 U/kg) but may point to insulin resistance if the insulin dosage is high (>1.5 U/kg). Some newer insulins (e.g., insulin glargine) have been developed to provide a more constant level of insulin, which should render the difference between the highest and lowest glucose level relatively small (so-called peakless insulins). In some pets, this is definitely the case; in most, however, insulin glargine is not a peakless insulin, and efficacy may be evaluated as mentioned earlier. In cases in which glucose concentrations are constantly high, insulin underdosing, insulin resistance, technical problems with the insulin administration procedure, stress hyperglycemia in cats, and the counterregulatory phase of the Somogyi phenomenon have to be considered.

After insulin effectiveness is confirmed, the glucose nadir should be interpreted; this should ideally be 5.0 to 7.8 mmol/L (90 to 140 mg/dl). A lower nadir can be seen with insulin overdosage, excessive overlap of insulin action, lack of food intake, and strenuous exercise. It is also typically seen in diabetic cats in which the disease is progressing to remission. In cases in which overdosage is identified, the insulin dosage should be reduced by 10% to 50%. Excessive overlap of insulin action may require a change to an insulin with a shorter duration of action. If the nadir is greater than 7.8 to 9 mmol/L (140 to 160 mg/dl) despite insulin effectiveness, insulin underdosage, stress hyperglycemia, the counterregulatory phase of the Somogyi phenomenon, and technical problems of the owners have to be considered. If the animal is already being treated with a high insulin dosage, insulin resistance also may be possible. It is very important to identify the exact cause because treatment decisions differ and may be completely opposite of one another. For example, insulin underdosage is treated by increasing the insulin dosage by 10% to 25%, whereas the Somogyi phenomenon is treated by decreasing the dosage by approximately 50%.

Careful evaluation of the other parameters of glycemic control such as clinical signs and serum fructosamine is mandatory. In addition, it may be necessary to assess the owner’s insulin administration technique. The counterregulatory phase of the Somogyi phenomenon may last 72 hours. Therefore several BGCs may be required to demonstrate a decrease of blood glucose to less than 3.5 mmol/L (60 mg/dl) followed by an increase in blood glucose to a concentration greater than 17 mmol/L (300 mg/dl) within a 12-hour period after insulin administration.

Duration of insulin effect is evaluated if the glucose nadir lies in the desired range of 5 to 7.8 mmol/L (90 to 140 mg/dl). The duration is defined as the time from insulin injection through the glucose nadir until the blood glucose exceeds 14 to 15 mmol/L (250 to 270 mg/dl). If the duration of insulin effect is less than 8 to 10 hours, animals usually show clinical signs of diabetes mellitus. In contrast, if the duration is longer than 14 hours, the risk of developing hypoglycemia secondary to overlapping insulin effect increases.

In some instances it is difficult to distinguish between too short duration of effect and the Somogyi phenomenon because BGCs may look alike. It is therefore important to reevaluate the patient soon (a few days to a week) after any treatment modification.