• Open Access

An Investigation of the Action of Neutral Protamine Hagedorn Human Analogue Insulin in Dogs with Naturally Occurring Diabetes Mellitus

Authors

  • C.A. Palm,

    1. Department of Clinical Studies, Matthew J. Ryan Veterinary Hospital, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA,
    Search for more papers by this author
  • R.C. Boston,

    1. Department of Biostatistics, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA
    Search for more papers by this author
  • K.R. Refsal,

    1. Animal Health Diagnostic Laboratory, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, MI
    2. Endocrine Section, Diagnostic Center for Population and Animal Health, College of Veterinary Medicine, Michigan State University, Lansing, MI
    Search for more papers by this author
  • R.S. Hess

    1. Department of Clinical Studies, Matthew J. Ryan Veterinary Hospital, School of Veterinary Medicine, University of Pennsylvania, Kennett Square, PA,
    Search for more papers by this author

  • Work was done at the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania (MJR-VHUP), School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA. Reported in part as an abstract in JVIM 21(3):595–596, 2007 (abstract # 86).

Corresponding author: Carrie A. Palm, 3850 Delancey Street, Philadelphia, PA 19104; e-mail: capalm77@gmail.com

Abstract

Background: Neutral Protamine Hagedorn human analogue insulin (Humulin N) is commonly used for treatment of canine diabetes mellitus (DM). However, blood glucose and serum insulin concentrations in Humulin N-treated dogs with naturally occurring DM have not been reported.

Objective: To investigate blood glucose and serum insulin concentrations in the clinical setting of client-owned Humulin N-treated dogs with naturally occurring, well-regulated DM.

Animals: Ten client-owned dogs with naturally occurring, well-regulated DM.

Methods: In this clinical study, blood glucose and serum insulin concentrations were measured when dogs received food and insulin (T0), at approximately every half hour for the next 2 hours, and then approximately every 2 hours for an additional 8 hours. Insulin duration of action was defined as the number of hours from T0 to the lowest blood glucose concentration and until blood glucose concentration returned to an interpolated value of 70% of basal blood glucose concentration (Glucoseb).

Results: Mean percent of insulin-induced blood glucose suppression was 49.9 ± 17.1% (median, 46%; range, 29–78%). Insulin duration of action ranged from 4 to 10 hours. Blood glucose concentration increased initially and returned to Glucoseb within 0.6–2.2 hours after T0 in 5 dogs. This initial blood glucose surge then was followed by blood glucose suppression in all 5 dogs.

Conclusions and Clinical Importance: These results suggest that Humulin N administered SC twice daily is an effective mode of treatment for dogs with naturally occurring DM. Postprandial hyperglycemia is present in some well-regulated diabetic dogs treated with Humulin N.

Neutral Protamine Hagedorn (NPH) human analogue insulin (Humulin Na) is commonly used for treatment of canine diabetes mellitus (DM).1,2 However, blood glucose and serum insulin concentrations in Humulin N-treated dogs with naturally occurring DM have not been reported. The goal of this study was to investigate blood glucose and serum insulin concentrations in a clinical setting of client-owned Humulin N-treated dogs with naturally occurring DM.

Materials and Methods

Ten dogs with well-regulated DM were enrolled in this clinical study. Dogs were determined to have well-regulated DM based on the presence and severity of clinical signs noted over a 4-week period before enrollment in the study, the pattern and dose of insulin therapy during this period, and physical examination findings. A CBC,b serum biochemistry,c fructosamine,d urinalysis,e and aerobic urine culture were performed in all dogs at the time of enrollment into the study. The time of enrollment into the study was defined as the time at which the investigation into the action of human analogue NPH insulin was performed.

A diagnosis of DM was confirmed by documenting persistent hyperglycemia and glucosuria. Diabetic dogs were included in the study if they were treated with an unchanged dose of twice daily SC human analogue NPH insulina that was ≤1.0 U/kg for at least 4 weeks before enrollment in the study.

Exclusion criteria were designed to exclude dogs with poor glycemic control. Exclusion criteria were clinical signs consistent with poor glycemic control such as polyuria, polydipsia, polyphagia, or weight loss, observed during the 4-week period before enrollment in the study, a change of insulin dose, administration of a dose of insulin >1.0 U/kg SC q12h within the 4-week period before enrollment into the study, or hypoglycemia (blood glucose concentration <50 mg/dLf) at any time during the 10-hour study period.

Intact female dogs and dogs receiving medications that could influence insulin action were excluded from the study. Medications warranting exclusion from the study included oral, topical, or ophthalmic corticosteroid-containing preparations. Dogs whose owners declined the request to sign the consent form were excluded from the study.

Body condition score (BCS) was assessed subjectively by 1 investigator (CAP) according to the following system: emaciated dogs were assigned a BCS of 1, very thin dogs were assigned a BCS of 2, thin dogs were assigned a BCS of 3, underweight dogs were assigned a BCS of 4, ideal weight dogs were assigned a BCS of 5, overweight dogs were assigned a BCS of 6, heavy dogs were assigned a BCS of 7, obese dogs were assigned a BCS of 8, and grossly obese dogs were assigned a BCS of 9.

University of Pennsylvania Institutional Animal Care and Use Committee (IACUC) approval was granted. In accordance with IACUC guidelines regarding blood sampling volume and body weight, only dogs with body weights ≥3.5 kg were included in the study.

Insulin action was investigated as follows: dogs were admitted to the hospital approximately 1 hour before the time that they routinely received food and insulin (T0). One dog that was reluctant to eat in the hospital was brought into the hospital immediately after eating its meal at home. Depending on owner compliance and time of admission, dogs had blood samples collected at 1 or 2 of the following time points: 60 minutes before T0 (T−60), 30 minutes before T0 (T−30), or immediately before T0. After T0, blood was collected from all dogs at approximately the same time intervals, over a period of 10 hours. During these 10 hours, blood was collected approximately every half hour for the 1st 2 hours (T0.5, T1, T1.5, T2) and then approximately every 2 hours for an additional 8 hours (T4, T6, T8, T10).

Basal insulin (Insulinb) and glucose (Glucoseb) concentrations were defined as the average of concentrations measured at T−30, and immediately before T0 in dogs in which these measurements were obtained, as the average of concentrations measured at T−60 and immediately before T0 in the dog in which these measurements were obtained, or as a single measurement obtained immediately before T0 or at T−60 in dogs in which this was the only measurement available.

One measure of the effect of insulin on blood glucose concentration over the entire 10-hour study period was determined by calculating the insulin-induced percent of blood glucose suppression. Percent of blood glucose suppression was calculated for each dog as illustrated in Figure 1. A blood glucose curve was plotted for each dog, and the area of interest was defined as the rectangular region between 2 parallel lines drawn through the value of a blood glucose concentration of 50 mg/dLf and through the point of maximal blood glucose concentration measured for each dog (just >360 mg/dL in the example provided in Fig 1). A concentration of 50 mg/dL was chosen retrospectively in order to allow inclusion of all measured glucose concentrations in the analysis. The lowest blood glucose concentration measured in the study was 53 mg/dL. The vertical lines of each rectangle were drawn through T0 and T10. For each dog, the area above the blood glucose curve (in light gray in Fig 1), divided by the entire area of the rectangle was defined as the insulin-induced percent suppression of blood glucose concentration. The area above and under the curve was determined using the trapezoidal rule.3

Figure 1.

 Illustration of parameters used for the calculation of insulin-induced percent of blood glucose suppression.

Another measure of the effect of insulin on blood glucose concentration over the entire 10-hour study period was determined by calculating insulin duration of action. Insulin duration of action was defined as the number of hours from T0 through the lowest blood glucose concentration and until blood glucose concentration returned to an interpolated value of 70% of Glucoseb.2,4 When blood glucose concentration decreased below 70% of Glucoseb but did not return to an interpolated value of 70% of Glucoseb within the 10-hour study period, insulin duration of action was defined as 10 hours. When blood glucose concentration did not decrease below 70% of Glucoseb duration of action was not calculated.

Blood glucose and serum insulin concentrations were measured at each time point. Blood glucose concentration was measured immediately after each blood collection with the Accu-Check Advantage portable blood glucose meter.f The blood glucose readings of this device recently have been shown to have excellent correlation with blood glucose concentrations determined by the hexokinase method, although blood glucose determinations using the Accu-Check Advantage device tend to be lower than those of the hexokinase method.g

Serum for insulin measurement was separated from whole blood within 20 minutes of each blood collection, or as soon as a blood clot formed. Serum samples for measurement of insulin concentrations were frozen at (−80 °C) and submitted to Veterinary Endocrine Laboratory, Michigan State University for analysis in 2 separate batches.

Insulin was measured in canine serum samples with a commercially available radioimmunoassay (RIA) kith utilizing a 2nd-antibody separation method. The RIA was performed with reagent volumes and incubation times specified in the manufacturer's protocol. The manufacturer of the kit described virtually no antibody cross-reactivity with other peptide hormones including glucagon, somatostatin, pancreatic polypeptide, or insulin-like growth factor I. The manufacturer also described cross-reactivity with human insulin (100%), porcine insulin (99%), and human proinsulin (55%). Aliquots of a canine serum pool (endogenous insulin concentration of 126 pmol/L) were spiked with human insulin to achieve increases of 36, 108, or 215 pmol/L. Recovery of these added increments of human insulin in the assay were 94, 102, and 116%, respectively. The sensitivity of the assay, defined as the concentration of insulin at 2 standard deviations (SD) from “0” specific binding was 15 pmol/L (mean of 10 assays). Canine serum samples with measured insulin concentrations of 282 and 788 pmol/L were diluted 50, 25, and 12.5% with “0” standard provided with the kit. When the results for the 282 pmol/L sample were corrected for dilution, 107, 105, and 110% recovery of the expected result was obtained for each of the respective dilutions. When results of the 788 pmol/L sample were corrected for dilution, 104, 94, and 85% of expected results were obtained for the 50, 25, and 12.5% dilutions, respectively. Intra-assay repeatability was assessed with 3 pools of canine sera assembled to have “low” (78 pmol/L), “mid-range” (316 pmol/L), and “high” (953 pmol/L) concentrations. The respective intra-assay coefficients of variation (CV) for 10 samples of each pool were 0.073, 0.038, and 0.048. In 10 assay runs, the interassay CV of canine serum pools having insulin concentrations of 90, 347, and 1,028 pmol/L were 0.150, 0.170, and 0.120, respectively.

Statistical Analysis

For data screening and exploration (baseline concentrations, nadir concentrations, and nadir times) we used tabulation, interpolation, graphical analysis, and trapezoidal integration (area under the curve determination).3 Normality of the data was confirmed by the Shapiro-Wilks test. To estimate disposition and appearance rates for insulin and glucose we used regression analysis (logarithmic in the case of disposition). To establish times at which insulin or glucose differed from baseline concentrations across the study population we used random effects regression methods with variation attributable to dog as the assigned random source. A probability level of .05 was used to distinguish between differences in observations attributable to chance (>.05) versus differences due to insulin action (<.05). All statistical analyses were performed by a statistical software package.i

Results

Mean (±SD) age of 10 dogs included in the study was 7.2 ± 1.8 years (median, 7 years; range, 5–11 years). Eight of the 10 dogs were neutered males and 2 were neutered females. Two dogs were of mixed breeding, 2 were Dachshunds, and there was 1 dog each of the following breeds: Bichon Frise, Pug, Yorkshire Terrier, Miniature Poodle, Labrador Retriever, and West Highland White Terrier.

Mean time between initial diagnosis of DM and enrollment in the study was 13.1 ± 12.5 months (median, 9 months; range, 1–36 months). The mean insulin dose administered for at least 4 weeks before inclusion in the study was 0.63 ± 0.18 U/kg SC q12h (median, 0.6 U/kg; range, 0.4–0.97 U/kg).

None of the dogs included in the study had polyuria, polydipsia, weight loss, or polyphagia within the 4-week period before enrollment, according to the owners.

Six dogs were fed Hill's W/Dj diet and 1 dog each was fed Hill's W/Dj with Eukanuba optimum weight control diet,k Eukanuba optimum weight control dietj with Iam's select bites with beef in gravy,l Nutro MAX weight control formula,m and Merrick senior medley dietn mixed with ALPO prime cuts diet in gravy with beef.o In addition to NPH insulin,a dogs received the following medications: Heartgardp (5 dogs), Frontlineq (4 dogs), Sentinelr (2 dogs), Advantixs (1 dog), Preventic Collart (1 dog), S-adenosyl methionine (denosyl)u (1 dog), Tramadolv (1 dog), Deramaxxw (1 dog), and famotidinex (1 dog).

Physical examination at the time of enrollment in the study was normal in 8 dogs. Two dogs had bilateral cataracts, and 1 dog each had overweight body condition, bilateral nuclear sclerosis, and multiple SC masses, or mild lip fold pyoderma and mild bilateral otitis. Cytologic evaluation of all SC masses was consistent with lipomas. Mean weight of all 10 dogs was 15.1 ± 11.7 kg (median, 9.4 kg; range, 5.3–41 kg). Mean BCS was 5.5 ± 1 of 9 (median, 5.0; range, 5–8). None of the dogs had thickening of the SC tissue in the region of insulin administration.

Insulinb and Glucoseb concentrations were calculated as the average of concentrations measured at T−30 and immediately before T0 in 7/10 dogs, as the average of concentrations measured at T−60 and immediately before T0 in 1 dog, or as a single measurement obtained immediately before T0 (1/10 dogs) or at T−60 (1/10 dogs).

Mean Insulinb for all 10 dogs was 131 ± 81 pmol/L (median, 111 pmol/L; range, 48–324 pmol/L, Fig 2). Mean maximum serum insulin concentration was 510 ± 261 pmol/L (median, 407 pmol/L; range, 199–1,064 pmol/L). The mean time to reach maximum serum insulin concentration was 1.6 ± 0.9 hours after T0 (median, 1.5 hours; range, 0.5–4.0 hours). Serum insulin concentration returned to Insulinb at a mean of 7.8 ± 2.4 hours after T0 (median, 8.5 hours; range, 3.1–10.0 hours). Four of 10 dogs had serum insulin concentrations that remained above Insulinb for the duration of the 10-hour study period. Mean serum insulin concentration at each time point was compared with Insulinb. Mean serum insulin concentrations at T0.5 (P= .035, 95% CI 7–200), T1 (P < .001, 95% CI 142–336), T1.5 (P < .001, 95% CI 266–459), T2 (P < .001, 95% CI 146–352), and T4 (P= .041, 95% CI 4–197) were significantly higher than mean Insulinb (Fig 2). Mean serum insulin concentrations at T6, T8, and T10 were not significantly different than mean Insulinb (Fig 2).

Figure 2.

 A plot of means (±SD) of Neutral Protamine Hagedorn insulina (pmol/L) concentrations in 10 dogs plotted over a 10-hour period. *Times at which the blood insulin concentration was significantly different from the baseline insulin (Insulinb).

Mean Glucoseb was 318 ± 135 mg/dL (median, 333 mg/dL; range, 110–563 mg/dL, Fig 3). Mean minimum blood glucose concentration was 128 ± 64 mg/dL (median, 112 mg/dL; range, 53–254 mg/dL). The mean time to reach minimum blood glucose concentration was 4.9 ± 3.0 hours after T0 (median, 4.0 hours; range, 1.0–10.0 hours).

Figure 3.

 A plot of mean (±SD) glucose concentrations (mg/dL) in 10 dogs receiving Neutral Protamine Hagedorn insulina plotted over a 10-hour period. *Times at which the blood glucose concentration was significantly different from the baseline glucose (Glucoseb).

Five dogs had an initial increase in blood glucose concentration shortly after insulin injection and feeding (T0). In these 5 dogs, blood glucose concentration returned to Glucoseb at 0.6, 0.6, 0.7, 1.6, and 2.2 hours after T0 (mean, 1.1±0.7 hours). This initial blood glucose increase then was followed by blood glucose suppression in all 5 dogs. Four of these 5 dogs were eating Hill's W/Dj only, and the 5th dog was eating Hill's W/Dj with Eukanuba optimum weight control diet.k

Mean blood glucose concentration at each time point was compared with Glucoseb. Mean blood glucose concentrations at T1.5 (P= .009, 95% CI 182–26), T2 (P < .001, 95% CI 238–72), T4 (P < .001, 95% CI 260–105), T6 (P= .001, 95% CI 213–58), T8 (P= .003, 95% CI 195–39), and T10 (P= .010, 95% CI 180–25) were significantly lower than mean Glucoseb (Fig 3). Mean blood glucose concentrations at T0.5 and T1 were not significantly different than mean Glucoseb (Fig 3).

Mean time from the time of mean maximum serum insulin concentration to the time of mean minimum blood glucose concentration was 3.3 ± 2.7 hours (median, 2.8 hours; range, 0.5–8.5 hours).

Mean percent of insulin-induced blood glucose suppression for all 10 dogs was 49.9 ± 17.1% (median, 46%; range, 29–78%). Insulin duration of action for all 10 dogs included in the study ranged from 4 to 10 hours. Mean insulin duration of action in 4 dogs in which duration of action could be calculated was 5.5 ± 1.9 hours (median, 5.0 hours; range, 4.0–8.0 hours). In 4 other dogs, blood glucose concentration decreased below 70% of Glucoseb but did not return to an interpolated value of 70% of Glucoseb within the 10-hour study period (as illustrated in Figs 4 and 5), and insulin duration of action was defined as 10 hours. In 2 additional dogs, blood glucose concentration did not decrease below 70% of Glucoseb and duration of action was not calculated.

Figure 4.

 Blood glucose concentration plotted against time in 1 well-regulated diabetic dog receiving food and Neutral Protamine Hagedorn insulina at T0. Blood glucose concentration ranged between 76 and 143 mg/dL over the 10-hour study period, and did not exceed 200 mg/dL.

Figure 5.

 Blood glucose concentration plotted against time in 1 well-regulated diabetic dog receiving food and Neutral Protamine Hagedorn insulina at T0. Baseline glucose (Glucoseb) was 350 mg/dL, and after the postprandial hyperglycemia, blood glucose concentration did not return to an interpolated value of 70% of Glucoseb. This dog had a 72% insulin-induced blood glucose suppression.

Discussion

Mean blood glucose concentrations became significantly lower than Glucoseb 1.5 hours after insulin injection and feeding, and remained significantly lower than Glucoseb for the duration of the study (until T10). These results suggest that human analogue NPH insulina administered SC twice daily is an effective mode of treatment for dogs with naturally occurring DM. These results support treatment of diabetic dogs with human analogue NPH insulina as an alternative to other insulin preparations that have been shown to be effective in treatment of canine DM.4–7

Mean serum insulin concentration was significantly higher than Insulinb 0.5 hour after insulin injection and feeding. This significant increase in serum insulin concentration occurred 1 hour before a significant decrease in mean blood glucose concentration was identified. One of the reasons for a time difference between the 1st significant increase in serum insulin concentration and the 1st significant decrease in mean blood glucose concentration may be postprandial hyperglycemia.

The presence of postprandial hyperglycemia was noted in 5 dogs in which glucose concentration returned to Glucoseb at a mean time of 1.1±0.7 hours after treatment with human analogue NPH insulina and feeding. All 5 dogs were fed a diet high in insoluble fiber content. This initial increase in blood glucose concentration then was followed by blood glucose suppression in all 5 dogs. Additional studies investigating specific therapeutic options for postprandial hyperglycemia in well-regulated diabetic dogs treated with human analogue NPH insulina and an appropriate diet may be needed to determine the clinical relevance of this finding.

Other studies of NPH insulin in dogs were designed to minimize the effects of food on insulin action by fasting study dogs.4,8 However, the present study was designed to ensure the findings were relevant to the clinical setting. Therefore, dogs were fed and given their insulin at the same time they routinely received food and insulin.

One of this study's limitations is that diet, exercise, and site of insulin injection were not standardized. These factors, along with potential stress hyperglycemia or presence of anti-insulin antibodies may have influenced serum insulin and blood glucose concentrations. Diet has been shown to influence glycemic control in dogs with naturally occurring DM.9 However, all of the dogs in this study were well regulated, and the 5 dogs that had postprandial hyperglycemia were fed a diet high in insoluble fiber as recommended for the management of canine DM.9 Other studies of dogs with well-regulated naturally occurring DM will be needed in order to determine whether and to what extent exercise, site of insulin injection, stress hyperglycemia, or presence of antihuman analogue NPH insulina antibodies may influence serial insulin and blood glucose measurements.

A time lag between the increase in insulin concentration and the decrease in glucose concentration was apparent when looking at the times at which mean maximum serum insulin concentration and mean minimum blood glucose concentration were reached. Mean time from the time of mean maximum serum insulin concentration to the time of mean minimum blood glucose concentration was 3.3 ± 2.7 hours (median, 2.8 hours; range, 0.5–8.5 hours). This may reflect the time required for insulin absorption, distribution, and action on blood glucose concentration in well-regulated dogs with naturally occurring DM.

Previous studies of the action of NPH insulin in diabetic dogs investigated NPH insulin preparations of porcine or bovine origin.4,8,10 These insulin preparations are no longer commercially available in the United States. The use of a human analogue NPH insulin in normal nondiabetic dogs has also been reported.10 The present study is the 1st report characterizing the action of human analogue NPH insulina in dogs with naturally occurring DM.

Results of this study suggest that when human analogue NPH insulina is administered to dogs with naturally occurring DM it has a pattern of peak development (Fig 2) as has been reported for this insulin product when it is administered to human beings.11

Various methods for calculation of insulin duration of action have been described in the veterinary literature.3,4 One commonly used definition of insulin duration of action is the time from insulin injection through the lowest blood glucose concentration and until blood glucose concentration exceeds 200–250 mg/dL.2 This definition of insulin duration of action is difficult to apply to many well-regulated diabetic dogs in which blood glucose concentration does not exceed 200 mg/dL, as illustrated in Figure 4, which depicts serial blood glucose measurements obtained in 1 of the dogs included in this study.

In the present study, insulin duration of action was defined as the number of hours from T0 through the lowest blood glucose concentration and until blood glucose concentration returned to an interpolated value of 70% of Glucoseb. This approach has been reported previously in the veterinary literature4 and suited the present study because permission for 12-hour blood glucose and insulin sampling could not be obtained, in accordance with IACUC guidelines regarding blood sampling volume.

The rationale for favoring an investigation of earlier blood insulin and glucose concentrations between T0 and T2 in favor of the 12-hour blood insulin and glucose sampling did give interesting and clinically relevant results with regard to postprandial hyperglycemia in dogs with naturally occurring well-regulated DM. This rationale was also based on the assumption that the 12-hour blood insulin and glucose concentrations would be similar to those measured at T0. Therefore, application of the previously described definition of duration of action utilizing a 70% value of Glucoseb at 10 hours in place of a 100% value of Glucoseb at 12 hours was reasonable. Another measure of the effect of insulin on blood glucose concentration over the entire 10-hour study period was applied by calculating the insulin-induced percent of blood glucose suppression. The percent of blood glucose suppression was calculated for each dog because, in 4 of the well-regulated diabetic dogs included in the study, blood glucose concentration did not return to an interpolated value of 70% of Glucoseb as illustrated in Figure 5 obtained for 1 of the study dogs. The dog depicted in Figure 5 had a 72% insulin-induced blood glucose suppression. One of the advantages of using the mean percent of insulin-induced blood glucose suppression as a measure of the effect of insulin on blood glucose concentration is that it can be applied to all dogs included in the study.

Future studies will be needed to determine if there is 1 optimal method of reporting the effect of insulin on blood glucose concentration in all dogs with naturally occurring DM, or whether it is best to apply several modes of analysis to the data in order to best describe it, as was done in the present study.

In 4 of the study dogs, blood glucose concentration did not return to an interpolated value of 70% of Glucoseb within the 10-hour study period, and insulin duration of action was defined as 10 hours. It is possible that in these dogs, duration of action of human analogue NPH insulina actually is longer than 10 hours, and that the mean insulin duration of action of analogue NPH insulina in dogs with naturally occurring diabetes is actually longer than the 7.75 ± 2.7 hours reported in this study. Other studies would be needed to determine the exact duration of action of human analogue NPH insulina in these dogs, and whether once daily treatment of some diabetic dogs with human analogue NPH insulina may be sufficient.

The SD bars of the mean values of insulin and glucose in Figures 2 and 3 indicate that there is individual variation in the action of human analogue NPH insulina among diabetic dogs that are clinically well regulated. This has been reported previously in poorly regulated diabetic dogs receiving a bovine NPH product and also is recognized in humans receiving human analogue NPH insulin.4,11 These results indicate that serial blood glucose and serum insulin measurements alone cannot be used to determine whether a diabetic dog is well regulated and that clinical signs must be considered for the assessment of glycemic control in diabetic dogs.12

Footnotes

aHumulin N, Eli Lilly and Co, Indianapolis, IN

bHematology analyzer, Celldyne 3500, Abbot Laboratories, Abbot Park, IL

cChemistry analyzer, Kodak Ektachem 250, Eastman Kodak Co, Rochester, NY

dFructosamine SPOTCHEM EZ, Heska Corporation, Loveland, CO

eUrinalysis N-Multistix SG, Bayer Corporation, Elkhart, IN

fAccu-Check Advantage, Roche Diagnostics, Indianapolis, IN

gCohen T, Nelson R, Kass P et al. Evaluations of six portable blood glucose meters in dogs. J Vet Intern Med 2008;22(3):729 (abstract #95)

hInsulin RIA, Diagnostic Systems Laboratories, Webster, TX

iIntercooled Stata 9.2 for Windows, Stata Corporation, College Station, TX

jW/D diet, Hill's Pet Nutrition Inc, Topeka, KS

kEukanuba Optimum Weight Control Diet, Proctor and Gamble Pet Care, Cincinnati, OH

lIams select bites with beef in gravy, Proctor and Gamble Pet Care

mNutro MAX weight control diet, Nutro Products Inc, City of Industry, CA

nMerrick senior medley diet, Merrick Pet Care, Amarillo, TX

oALPO prime cuts diet in gravy with beef, Nestle Purina Pet Care Company, St Louis, MS

pHeartgard, Merial, Duluth, GA

qFrontline Plus, Merial

rSentinel, Novartis Animal Health Incorporated, Greensboro, NC

sAdvantix, Bayer, Shawnee Mission, KS

tPreventic Collar, Virbac Animal Health Incorporated, Fort Worth, TX

uS-adenosyl Methionine (Denosyl), Nutramaxx Laboratories Incorporated, Edgewood, MD

vTramadol, Caraco Pharmaceutical Laboratories, Detroit, MI

wDeramaxx, Novartis Animal Health Incorporated, Basel, Switzerland

xFamotidine, Major Pharmaceuticals, Livonia, MI

Acknowledgments

This work was supported by a grant from the Department of Clinical Studies, MJR-VHUP, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, PA and by a gift from the Gruenloh family in memory of Kiri Gruenloh, a beloved diabetic dog.

Ancillary