Materials and Methods

Boxers and Doberman Pinschers evaluated at The Ohio State University College of Veterinary Medicine between January 2003 and November 2004 were eligible for the study if they had been evaluated via 24-hour ambulatory electrocardiography and had a plasma sample available for analysis. All blood samples had been centrifuged, and the plasma was frozen at −80°C until analysis. Standard 2-dimensional, M-mode, and color flow Doppler echocardiography was performed.^a A 3-channel transthoracic ambulatory electrocardiography system was attached to each dog for 25 hours.^b A technician under the guidance of a veterinary cardiologist analyzed ambulatory electrocardiography recordings for the total number of VPCs/24 h by use of an analysis system.^c Any dogs with recordings that did not have at least 20 hours of readable data were not included in the study. Dogs were considered to be affected by ARVC or DCM if they had > 100 VPCs/24 h or a fractional shortening of < 25%. Dogs receiving antiarrhythmic medications or fish oil supplements were excluded from analysis.

Lipid extraction—Total phospholipid fatty acids were analyzed.^d Plasma lipids were extracted according to the method of Bligh and Dyer,¹² whereby mixtures of plasma, methanol, chloroform, and water are prepared such that lipid is recovered in a chloroform layer. Trace amounts of butylated hydroxyl toluene (0.05%) were added to the chloroform to prevent oxidation of polyunsaturated fatty acids. The resulting lipid extracts were maintained under an atmosphere of nitrogen following extraction and kept frozen prior to additional processing.

Separation of lipid classes—Immediately prior to lipid class separation, lipid samples were dried under a gentle stream of nitrogen, rediluted in 50 mL of chloroform, and prepared for lipid class separation.

Lipid classes including total phospholipid, nonesterified fatty acids, triglyceride, and cholesterol esters were separated on commercial silica gel G plates.^e The chromatographic plates were developed in a solvent system consisting of distilled petroleum ether (boiling point, 30° to 60°C):diethyl ether:acetic acid (80:20:1, by volume). Following development, the silica gel plates were sprayed with a methanolic solution containing 0.5% 2,7-dichlorofluorescein, which was used to visualize lipid classes under UV light. Desired corresponding lipid bands were scraped into screw-cap tubes. The samples were transesterified with boron trifluoride (12%) in methanol^f in an 80°C water bath for 90 minutes. Resulting fatty acid methyl esters were extracted with water and petroleum ether and stored frozen at −80°C until gas chromatographic analysis was performed.

Fatty acid analysis—Lipid class fatty acid methyl ester composition was determined by use of capillary gas chromatography. Methyl ester samples were dessicated to dryness under nitrogen and resuspended in heptane containing methyl-tridecanoic acid^g as an internal standard. Resulting fatty acid methyl esters were separated and quantified with a gas chromatograph^h by use of a 30-m free fatty acid–phase coating and commercial software.ⁱ The instrument temperature was programmed from 190° to 240°C at 7°C/min, with a final hold of 10 minutes, which separated and measured fatty acid methyl esters ranging from 12:0 to 24:1. The detector temperature was 250°C. Helium carrier gas was used at a flow rate of 1.4 mL/min and a split ratio of 1:25. Chromatographic data were collected and processed with commercial software.ⁱ Individual peaks, representing as little as 0.05% of the fatty acid methyl esters, were distinguished.

Fatty acids were identified by comparison to authentic fatty acid standards and quantitated with peak area and internal standard. Resulting data are expressed as percentage composition.

Statistical analysis—Data were examined graphically before analysis. All data are reported as median (range) for consistency because many of the fatty acid concentrations were not normally distributed. Data that were not normally distributed were transformed prior to analysis, and independent t tests were used to compare subject characteristics and plasma fatty acid concentrations between the 2 breeds, whereas Pearson correlations were used to compare individual plasma fatty acid concentrations with the number of VPCs/24 h. Commercial statistical software was used for analysis.^j For all analyses, P < 0.05 was considered significant.

Results

Fifty-one dogs (38 Boxers and 13 Doberman Pinschers) were enrolled in the study (Table 1). Median age, sex, and number of VPCs/24 h were not significantly different between the 2 breeds. Echocardiograms were performed on 49 of 51 dogs. Two Doberman Pinschers and 1 Boxer had fractional shortening < 25% (P = 0.10). Seventeen dogs (13 Boxers and 4 Doberman Pinschers) were considered affected by either ARVC or DCM. Medications administered to the dogs included digoxin (1 Boxer), phenylpropanolamine (1 Boxer), phenobarbital (1 Boxer), and thyroxine (2 Boxers and 3 Doberman Pinschers). Some dogs were receiving more than 1 medication.

Table 1—

Signalment and plasma fatty acid concentrations (median [range]) of 38 Boxers and 13 Doberman Pinschers. Plasma fatty acid concentrations reported as percentage normalized values.

View Table

Compared with the Doberman Pinschers, Boxers had significantly (P < 0.001) higher plasma concentrations of γ-linolenic acid (C18:3W6) but had lower (P = 0.005) concentrations of arachidonic acid (C20:4 n-6;). Plasma concentrations of total polyunsaturated fatty acids (P < 0.001) and total n-6 fatty acids (P < 0.001) were higher in the Doberman Pinschers. Plasma concentrations of other major fatty acids were not significantly different between the 2 breeds. If only the 17 affected dogs were compared, only plasma γ-linolenic acid concentration was significantly (P = 0.03) different between the 13 Boxers and 4 Doberman Pinschers.

For all 51 dogs as a group, there were no significant correlations between plasma fatty acid concentrations and VPC frequency. However, in Boxers, but not in Doberman Pinschers, concentrations of oleic acid (C18:1 n-9; r = 0.37; P = 0.02), total n-9 fatty acids (r = 0.37; P = 0.02), and total n-3 fatty acids (r = 0.37; P = 0.02) were positively, but weakly, correlated with VPC frequency. If only the 17 affected dogs were considered (13 Boxers and 4 Doberman Pinschers), plasma total n-3 fatty acid concentrations were significantly correlated with VPC frequency (r = 0.52; P = 0.03).

Discussion

Results of this study suggested that some plasma fatty acid concentrations were different between the Doberman Pinschers and Boxers, both at the level of certain individual fatty acids (eg, γ-linolenic acid and arachidonic acid) and at the level of specific types of fatty acids (eg, total n-6 fatty acids and total polyunsaturated fatty acids). Although differences in plasma and myocardial fatty acids have been reported for dogs with DCM, compared with healthy controls, breed-specific differences in plasma fatty acid concentrations have not been previously reported.^6,7

In addition to plasma fatty acid concentration differences, the relationships between types of fatty acids and VPC frequency may have differed between dogs of the 2 breeds in this study, although further investigation is required. In Boxers, concentrations of total n-3 fatty acids, oleic acid (C18:1 n-9), and total n-9 fatty acids were positively, although not strongly, correlated with VPC frequency, whereas in Doberman Pinschers, there were no significant correlations. When only the 17 affected dogs were included, there was a significant correlation between plasma total n-3 fatty acid concentrations and VPC frequency for both breeds combined. A lack of correlation in the Doberman Pinschers may have been attributable to the small number of dogs of this breed. The weak positive correlations between n-3 fatty acids and arrhythmia reported in the present study differed from results in earlier studies^{8,10,11,13,14} of experimental animals and humans, in which n-3 fatty acids have been reported to confer protection from ventricular arrhythmia. Notably, in a recent clinical trial limited to Boxers with ARVC, supplementation with EPA and DHA reduced VPC frequency, and n-3 fatty acid concentration was negatively correlated with percentage change in VPC frequency in dogs that received EPA and DHA.¹¹ The negative correlation in that study between n-3 fatty acids and change in VPC frequency detected in dogs that received EPA and DHA suggested to the authors that a minimum threshold concentration of n-3 fatty acids might be required for an effect on VPC frequency to be detected.¹¹ In the present study, median concentrations of total n-3 were substantially lower in Boxers and Doberman Pinschers (3.4% and 3.0%, respectively) than concentrations in dogs that received EPA and DHA in the previous clinical trial in Boxers with ARVC.¹¹ The lack of a negative correlation between n-3 fatty acid concentration and frequency of VPCs in light of the relatively low n-3 fatty acid concentrations in the present study may support the hypothesis that a threshold concentration of n-3 fatty acid is needed to affect frequency of VPCs. Compared with n-3 fatty acids, relationships between n-9 fatty acids and arrhythmias have been much less studied. One study¹⁵ evaluating antiarrhythmic effects of various fatty acids in rats not only failed to find an antiarrhythmic effect for oleic acid (C18:1 n-9) but also did not detect a positive correlation.

The reasons for the breed-based differences in plasma fatty acid concentrations in the present study are not clear, but may be related to metabolic differences between Doberman Pinschers and Boxers and differences in the underlying cardiomyopathies. The histologic lesions typically associated with the 2 forms of cardiomyopathy overlap in some respects and differ in others. Although myocyte loss and fibrous replacement occur with both forms of disease, the adipocyte replacement of cardiomyocytes that characterizes Boxers with ARVC has not consistently been documented in Doberman Pinschers with DCM.^3,16 Another difference between the cardiomyopathies in the 2 breeds is that, in Boxers, ARVC lesions tend to originate or predominate in the right ventricle, in contrast with the left-sided or biventricular changes that characterize DCM in Doberman Pinschers.^2,16,17 The extensive cardiac remodeling associated with DCM may have greater implications for systolic function than does ARVC. In the present study, fractional shortening was reduced to < 25% in 2 of 13 Dobermans Pinschers but in only 1 of 38 Boxers; however, caution in interpretation is necessary because of the small number of dogs with reduced fractional shortening. The metabolic consequences of reduced systolic function in DCM may differ from the right-sided cardiomyopathic lesions of ARVC that may cause a high frequency of arrhythmia without severely impairing cardiac function.

There are several limitations to this study that could have affected outcomes and complicated the interpretation of results. Because detailed diet histories were not available, intakes of n-3 fatty acids were unknown and plasma n-3 fatty acid concentrations were highly variable, especially in the Boxers. In addition, 5 dogs were receiving thyroxine, which at higher doses can increase susceptibility to arrhythmias. It is not known how many of the dogs that were not considered affected by ARVC or DCM will eventually develop cardiomyopathy, and this may complicate the interpretation of fatty acid differences and correlations. Finally, the cutoff value to define dogs as affected with ARVC is controversial in the veterinary cardiology community. In the present study, we chose to use > 100 VPCs/24 h as the cutoff value to consider a dog affected, but further work will be needed to establish whether this value is the most appropriate. Future studies are needed to confirm these results, to determine whether fatty acid metabolism differs in Boxers and Doberman Pinschers, and to evaluate the effect of fatty acid intake on arrhythmia frequency in Doberman Pinschers.