Effect of a low‐fat diet on serum triglyceride and cholesterol concentrations and lipoprotein profiles in Miniature Schnauzers with hypertriglyceridemia

Abstract Background Hypertriglyceridemia is common in Miniature Schnauzer (MS). Dietary management of hypertriglyceridemia is important, but no studies are available. Hypothesis/Objectives To evaluate the effect of a commercially available low‐fat diet on serum triglyceride and cholesterol concentrations and lipoprotein profiles in MS with hypertriglyceridemia. Animals Sixteen MS with hypertriglyceridemia and 28 MS without hypertriglyceridemia. Methods Prospective clinical trial. Four blood samples (1‐2 months before and 1 day before diet change and 2 and 3 months after the dogs were fed the low‐fat diet) were collected from the MS with hypertriglyceridemia. Results Serum triglyceride concentrations for the 2 samples after the diet change (median of sample 3 = 177 mg/dL; range, 48‐498; median of sample 4 = 168 mg/dL; range, 77‐745) were significantly lower than the 2 samples before the diet change (median of sample 1 = 480 mg/dL; range, 181‐1320; median of sample 2 = 493 mg/dL; range, 114‐1395; P < .001). Serum cholesterol concentrations for the 2 samples after the diet change (mean for sample 3 = 257 mg/dL, SD = 82.2; mean for sample 4 = 178 mg/dL, SD = 87.4) were also significantly lower than the 2 samples before the diet change (mean for sample 1 = 381 mg/dL, SD = 146.1; mean for sample 2 = 380 mg/dL, SD = 134.7; P < .001). Before the diet change, 15/16 (94%) of hyperlipidemic MS were classified as hyperlipidemic based on their lipoprotein profiles alone. After the diet change, significantly fewer MS (7/16; 44%; odds ratio = 19.3; 95% CI = 2.0‐184.0; P = .006) were classified as hyperlipidemic based on lipoprotein profile analysis. Conclusions and Clinical Importance The study diet was effective in reducing serum triglyceride and cholesterol concentrations and correcting lipoprotein profiles in MS with hypertriglyceridemia.

Conclusions and Clinical Importance: The study diet was effective in reducing serum triglyceride and cholesterol concentrations and correcting lipoprotein profiles in MS with hypertriglyceridemia.

| INTRODUCTION
Idiopathic or primary hyperlipidemias in dogs are a group of metabolic disorders of diverse etiology that are more common in certain breeds. 1 The phenotypic appearance of these disorders as determined by routine clinicopathological testing is quite uniform and typically characterized by increased serum concentrations of triglycerides, cholesterol, or both. However, more in-depth analysis of these lipid abnormalities, by lipoprotein profiling, often reveals important differences among breeds. 2 Primary hypertriglyceridemia was first described in the Miniature Schnauzer (MS) and occurs in dogs of this breed in the United States and several other countries. [3][4][5] Hypertriglyceridemia is present in 33% of 192 MS from the United States investigated, with >75% of dogs over 9 years of age being affected, suggesting that it is possibly the most common lipid abnormality in dogs. 4 Hypertriglyceridemia in MS is typically characterized by increases of various degrees in very low density lipoproteins (VLDL) and chylomicrons, with or without concurrent hypercholesterolemia. 2,6 Several pieces of evidence suggest that hypertriglyceridemia in MS is a condition of major clinical importance. While many animals are asymptomatic, 4 hypertriglyceridemia in this breed has been linked to other disorders such as hepatobiliary disease (eg, gall-bladder mucocele, vacuolar hepatopathy), 7,8 pancreatitis, 9,10 insulin resistance, 11 glomerular disease (eg, glomerular lipidosis), 12,13 ocular disease (eg, ocular lipid deposits, lipemic uveitis), 14,15 and neurological abnormalities (eg, seizures). 1 We have recently reported a detailed analysis of the lipoprotein profiles of MS with hypertriglyceridemia using a novel lipoprotein density profiling method. 2 However, the biochemical, metabolic, and genetic bases of hypertriglyceridemia in MS are yet to be elucidated. Therefore, definitive recommendations for effective prevention and specific management of this condition are currently unavailable. Due to the serious and potentially even fatal diseases that have been associated with hypertriglyceridemia in MS, management of hypertriglyceridemia is mandatory even in dogs in which clinical signs are not present. 1 The most commonly recommended initial approach in the management of hypertriglyceridemia is the use of a low-fat diet. 1 There are several commercially available diets marketed as Miniature Schnauzers with hypertriglyceridemia of various degrees were included in this group. These dogs were selected from a pool of >300 MS that were enrolled as part of several ongoing projects related to hypertriglyceridemia in this breed. The requirements for inclusion of dogs in group 1 in the present study were: (a) the presence of hypertriglyceridemia (dogs with combined hypertriglyceridemia and hypercholesterolemia were also included in the study if all other requirements were met; dogs with hypercholesterolemia alone were excluded); (b) consuming diets that were not labeled as "low-fat"; (c) absence of any clinical signs at the time of initial blood collection; (d) no history of diseases or current use of drugs known to affect lipid metabolism 1 ; (e) a body condition score (BCS), between 4 and 6 on a scale of 1 to 9 (the BCS was assessed by the referring veterinarian using a printed guide with pictures and instructions that was provided to each referring veterinarian); and (f) willingness of each dog's owners to enroll their dog in the study.

| Study design
Each 1 of the dogs in group 1 had a total of 4 blood samples collected. The first sample (sample 1) was used to investigate the presence of hypertriglyceridemia and select which dogs would be candidates for enrollment into the study. Dogs with hypertriglyceridemia that also met all the criteria for enrollment mentioned above were chosen to continue with the study. In order to confirm the results of the initial sample and to investigate the variability of hypertriglyceridemia over time, a second blood sample (sample 2) was collected 1 to 2 months after the collection of the initial blood sample. During that time, the owners were instructed to not make any changes to the diet type or amount, or the activity of their dogs. If hypertriglyceridemia was confirmed with the second sample, the dogs were placed on the study diet (see below).
Owners were instructed to gradually switch their dogs from the original diet to the study diet over 7 days. Because all dogs had a normal BCS, the amount of diet was determined based on the normal weight guide on the package of the diet. In addition, the owners were instructed to weigh their dogs once a week and adjust the amount of food accordingly in order to maintain a relatively stable body weight for the duration of the study. Additional instructions given to the owners involved ensuring that the study diet was fed exclusively to the dogs enrolled in the study, with the exception of small quantities of steamed vegetables (carrots and broccoli) that could be used as treats. No lipid-lowering medications were used in any of the dogs for the duration of the study. Approximately 7 to 9 weeks after the dogs had been exclusively on the study diet, a third blood sample (sample 3) was collected.
Finally, a fourth sample (sample 4) was collected approximately 2 to 4 weeks after the third sample, and while the dogs were still exclusively on the study diet. This was done in order to confirm the results of the 3rd sample and to investigate the variability of serum triglyceride and cholesterol concentrations over time while the dogs were on the low-fat diet.
Serum triglyceride and cholesterol concentrations were measured in all samples and compared between time-points. Two cut-off points were used for categorical analysis of serum triglyceride concentrations before and after diet change. The first cut-off point was the upper limit of the reference interval of the assay (108 mg/dL; see below).
However, increases in serum triglyceride concentrations that are slightly above the reference interval are considered mild and might not be associated with any appreciable risk for complications. In addition, there is anecdotal evidence of age-related increases in serum triglyceride and cholesterol concentrations in dogs other than Miniature Schnauzers. Therefore, a second cut-off point (500 mg/dL) was selected to indicate dogs that are considered to have severe hypertriglyceridemia and possessed of an appreciable risk for complications.
The value of 500 mg/dL was selected as a value commonly used in clinical practice as indicative of extreme hypertriglyceridemia; there is no evidence-based cut-off for serum triglyceride concentrations that indicates an increased risk for all conditions associated with hypertriglyceridemia. Lipoprotein profiles were determined as previously described 2 and comparisons made between time-points (samples [1][2][3][4] and between groups (hyperlipidemic vs. nonhyperlipidemic).
Group 2 dogs were enrolled on a 1-time basis and no follow-up samples were collected from these dogs. As mentioned above, these dogs were only used to provide normative lipoprotein profile data for apparently healthy MS. Serum triglyceride, cholesterol, and Spec cPL concentrations were also measured in all samples of dogs in this group but were not used for any comparison with group 1 (see above).

| Study diet
The diet selected for the present study was a commercially available therapeutic diet labeled as "low-fat" in dry form (Royal Canin Gastrointestinal Low-fat, Royal Canin USA, Inc, St. Charles, Missouri). The fat content of the study diet was 18.6 g of fat per 1000 Kcal.

| Blood collection and handling
Owners that agreed to participate in the study were each sent a styrofoam box containing ice packs and the material necessary for blood collection and were asked to schedule an appointment with their veterinarian for the blood collection. All dogs were required to be fasted for a minimum of 12 hours before each scheduled blood collection.
Ten milliliters of blood were collected from each dog into a red-top tube (with no additive). Blood samples were centrifuged immediately after clot formation and the serum separated from the clot. Serum samples were sent to the Gastrointestinal Laboratory packed on ice by overnight courier. Serum samples were stored at −80 C until further use.

| Standardized questionnaire
Owners and/or primary care veterinarians of all dogs were asked to complete a standardized questionnaire for each dog and for each time a blood sample was collected. Questions covered date of birth, sex, body weight, BCS, current diet(s), current medications, and current and past health history of the dogs. Questionnaires from all dogs were reviewed to determine whether the dogs fit the inclusion criteria for each group and whether any of the variables evaluated changed during the study (eg, body weight or BCS).

| Ethics approval
The owners of each dog enrolled in the study had to sign an informed owner consent form. The study protocol was reviewed and approved by the Clinical Research Review Committee at Texas A&M University (TAMU-CRRC# 2008-37).

| Lipoprotein profile analysis
Lipoprotein profiling was carried out using a bismuth sodium ethylenediaminetetraacetic acid (NaBiEDTA) density gradient ultracentrifugation method as previously described. 2,16 The sodium salt of BiEDTA has been described as a novel solute forming a selfgenerating density gradient during ultracentrifugation of serum samples for the separation of lipoproteins. 16 Briefly, for each sample,  Normally distributed data were analyzed using t tests, while non-normally distributed data were analyzed using Mann-Whitney tests. For multiple comparisons, repeated measures 1-way analysis of variance (ANOVA) was used for normally distributed paired data followed by Bonferroni analysis, while the Friedman's test was used for non-normally distributed paired data followed by Dunn's post hoc multiple comparison tests. Proportions were compared between groups using Fisher's exact tests with calculation of the odds ratio (OR) and the 95% confidence interval (95% CI) to test the possibility of an association between categorical variables.
Sliced inverse regression (SIR) or logistic regression analysis was used to test whether there was a relationship between group (diet, time-point) and lipoprotein profiles. The SIR provides a linear discriminant analysis (LDA) value that ranks individuals within the group and generates graphical plots to show separation in groups.
Significance was set at P < .05 for all analyses.

Patients characteristics
Age (y), median (range) 8.5 (6.7-11.9) 9.1 (7. Data on the diets the dogs were on before the diet change were collected. However, these data were impossible to analyze either because the dogs were on several different commercial diets at the same time or because they were being fed a combination of commercial diets and home-made diets. As mentioned above, none of the commercial diets were labeled as "low-fat' and the owners were instructed to not change the amount or type(s) of diet or the dog's activity and lifestyle for the duration of the trial (with the exception of the study diet). F I G U R E 1 A, Serum triglyceride and, B, cholesterol concentrations in dogs with hypertriglyceridemia before and after diet change. Samples 1 and 2 were collected while dogs were on their original diets. Samples 3 and 4 were collected approximately 8 and 12 weeks, respectively, after the dogs were placed on the study diet. There was a significant decrease in serum triglyceride (P < .001) and cholesterol (P < .001) concentrations after the diet change. The dotted lines represent the upper limit of the reference interval. See also Tables 4 and 5 Group 2 consisted of 28 Miniature Schnauzers that had a median BCS of 5 (range, 4-6) and a median age of 9.1 years (range, 7.1-12.2 years). Fourteen dogs were female (8 spayed) and 14 dogs were male (5 castrated).  Figure 1A; Table 3). Tukey's post hoc test indicated that there were no significant differences between samples 1 and 2 (ie, while dogs were on their original diets) or between sample 3 and 4 (ie, while dogs were on the study diet). There were significant differences between samples before and after the diet change (P < .05 for all; Table 4). Furthermore, significantly more dogs had P < .001; Figure 1B; Table 3). Post-ANOVA analysis using Bonferroni's correction for multiple comparisons indicated that there were significant differences between individual comparisons before and after low-fat diet consumption, with the exception of samples 2 and 4, which approached but did not reach statistical significance ( Figure 1B; Tables 3 and 5). However, there was no significant difference in the proportion of dogs that had hypercholesterolemia before (8/16) and after the diet change (3/16; P = .14).  fractions (mainly LDL 1 and LDL 2 ) and higher HDL fractions (mainly HDL 2a , HDL 3b , and HDL 3c ) than responders ( Table 2).

T A B L E 3 Serum triglyceride and cholesterol concentrations in Miniature Schnauzers with idiopathic hypertriglyceridemia (n = 16) before (samples 1 and 2) and after (samples 3 and 4) dietary intervention
The most important changes in the lipoprotein density distribution of dogs in group 1 in response to diet change were decreases in TRL and LDL 1 and, to a lesser degree, increases in LDL 4 and HDL 3c (Table 6). Figure 4A,B shows the lipoprotein density profiles from a representative dog in group 1 before and after the diet change. Note: Mean integrated intensities before and after diet change as well as the percent of changes are displayed.
differences were detected only when serum triglyceride and cholesterol concentrations were compared between time-points before and after the diet change clearly suggests that this was due to an effect of the diet.
An interesting and unexpected finding of our study was the great variation of serum triglyceride concentrations over time within individual dogs and while consuming the same diet. Figure 1A clearly  formation depends upon dietary fat consumption and hyperchylomicronemia is expected to respond to low-fat diets. 17 In contrast, increases in VLDL, which are also triglyceride-rich and isolated in the TRL fraction with our methodology, are not necessarily responsive to low-fat diets because VLDLs are produced through the endogenous pathway and contain triglycerides derived from de novo lipogenesis. 17 Our study supports the notion that lipoprotein profile