The Relationship of Resting S-T Segment Depression to the Severity of Subvalvular Aortic Stenosis and the Presence of Ventricular Premature Complexes in the Dog

Statistical Analysis

Descriptive statistics were performed to evaluate PG, age, ECG, HR, number of VPCs per 24 hours, and STDV for all of the dogs that met the inclusion criteria of the study. A Pearson’s product-moment correlation coefficient^b was performed to determine whether STDV correlated with PG, age, HR, and number of VPCs in a 24-hour period.

Descriptive statistics were performed to evaluate PG, age, HR, and number of VPCs per 24 hours for the population of dogs with STD and the population of dogs without STD. Student’s t-test was used to compare the PG, age, HR, and number of VPCs per 24 hours in the dogs with STD to the PG, age, HR, and number of VPCs per 24 hours in the dogs without STD. If the values were not normally distributed, a Mann-Whitney rank sum test was performed. A P value of ≤0.05 was considered significant.

The sensitivity, specificity, and positive and negative predictive values of STD were determined for the group of dogs with a PG ≥80 mm Hg, since this gradient has been previously used to define severe SAS.³ Sensitivity was calculated by dividing the number of dogs that had STD and a PG ≥80 mm Hg by the total number of dogs that had ≥80 mm Hg. Specificity was determined by dividing the number of dogs without STD and with a PG <80 mm Hg by the total number of dogs with PG <80 mm Hg. Positive predictive value was determined by dividing the number of dogs with STD and a PG ≥80 mm Hg by the total number of dogs with STD. Negative predictive value was determined by dividing the number of dogs without STD and with a PG <80 mm Hg by the total number of dogs without STD.

Results

A total of 35 dogs with a maximal instantaneous PG ≥50 mm Hg had both ECG and AECG performed. The breeds of dogs included Irish wolfhound (n=1), German shepherd dog (n=1), rottweiler (n=1), boxer (n=5), golden retriever (n=7), and Newfoundland (n=20). Ages ranged from 4 to 27 months (median, 9 months). The maximal instantaneous PG ranged from 50 to 228 mm Hg (mean, 119 mm Hg). The HR on the ECG ranged from 60 to 200 beats per minute (BPM; mean, 109 BPM). The STDV on the ECG ranged from 0 to 0.6 mV (median, 0.1 mV). The number of VPCs per 24 hours on the AECG ranged from 0 to 11,124 (median, 2 VPCs). A statistically significant correlation was not found between STDV and PG, age, HR, or number of VPCs per 24 hours for the entire study population.

A total of 22 dogs did not have STD. The age range for this group of dogs was 5 to 27 months (median, 10 months). Maximal instantaneous PG ranged from 50 to 228 mm Hg (mean, 112 mm Hg). The HR on the ECG ranged from 78 to 200 BPM (median, 108 BPM). The number of VPCs per 24 hours on the AECG ranged from 0 to 11,124 (median, 3 VPCs).

A total of 13 dogs demonstrated STD. Age ranged from 4 to 20 months (median, 9 months). Maximal instantaneous PG ranged from 70 to 215 mm Hg (mean, 134 mm Hg). The HR on ECG ranged from 60 to 140 BPM (mean, 106 BPM). The number of VPCs per 24 hours on the AECG ranged from 0 to 3,680 (median, 3 VPCs).

A statistical difference was not observed for PG, age, HR, or number of VPCs per 24 hours for the group of dogs with STD when compared to those without STD.

Twenty-nine dogs had a PG ≥80 mm Hg. The sensitivity of STD for a PG ≥80 mm Hg was 63%, and the specificity was 83%. The positive predictive value was 92%, with a negative predictive value of 22%.

Discussion

In the study presented here, a significant difference in the severity of SAS (defined by PG), age of dog, HR, or number of VPCs per 24 hours was not found between the group of dogs with STD and the group without STD. Additionally, a correlation was not observed between STDV and these variables for the population as a whole. For purposes of analysis, the authors defined STD as ≥0.2 mV based on a published value for dogs.¹¹ Although this criterion did not identify differences in PG or the number of VPCs for dogs with STD as compared to dogs without, it was found to be reasonably specific for identification of dogs with a PG of ≥ 80 mm Hg.

Overall, only 37% (13/35) of the population studied had STD as it was defined. The S-T segment depression is believed to be an indicator of subendocardial ischemia and has been reported in both dogs and humans with severe SAS and presumed myocardial ischemia.⁴¹²¹³ The inability to detect STD in a larger percentage of the population could be related to several possible issues, including the population studied, the presence or absence of ischemia in dogs with SAS, or an inability to accurately assess STD on an ECG. First, the study might have been biased in a way that dogs unlikely to have STD were selected. It is possible that the dogs selected were not severely affected enough to produce significant myocardial ischemia and subsequent STD; however, 83% (29/35) of the dogs met the previously defined criteria for severe SAS that has been shown to identify dogs at increased risk for sudden cardiac death.³ If sudden cardiac death is associated with ventricular arrhythmias and myocardial ischemia, as has been suggested, the authors would have expected dogs with this level of severity to be more likely to show STD.³⁴ It is possible that the development of ventricular hypertrophy, myocardial ischemia, and STD may be related to the length of time that the ventricular myocardium is exposed to the elevated ventricular pressure. This might suggest that older animals would be more likely to have STDV. However, a correlation between age and STDV was not observed in this study. It should be noted, though, that the population of dogs evaluated was fairly immature (median, 9 months), and further studies need to be performed with older dogs to fully understand the relationship between PG, age, and STDV. An alternative way to interpret this data may be to suggest that STD was not detected more frequently because myocardial ischemia was not present, and that dogs with moderate to severe SAS do not suffer from significant myocardial ischemia. This would seem contrary to previous reports regarding the presence of myocardial ischemia as identified by an increase in hyperechogenicity by echocardiography in dogs with severe SAS.¹³ Additionally, support for the presence of subendocardial ischemia has been based on the theory that ischemia may develop, at least in part, from the reversal of coronary artery systolic flow that occurs in dogs with SAS.¹⁴ However, it is possible that the degree of myocardial ischemia that develops with SAS in the dog is less than that in the human patient because of the substantial collateral circulation seen in the dog.¹⁵ Postmortem samples on these dogs are not available; therefore, the authors cannot determine the presence or absence of ischemic changes. A more likely theory may be that myocardial ischemia was present more frequently than suggested by the presence of STD, but that the method of assessment of STD used was not accurate. Evaluation of STD in lead II of the limb leads may be an insensitive method for assessing myocardial ischemia in the dog. In some human studies that evaluate STD, the precordial leads are used preferentially.⁷¹⁶¹⁷ The evaluation of the V6 lead for STD in the dog has been suggested by some authors.¹¹ Unfortunately, this lead is rarely evaluated in routine clinical practice and was not obtained for the dogs used in this study. Additionally, STD might have been more easily detected if the dogs had been evaluated immediately after exercise. Systolic coronary artery flow reversal in dogs with SAS has been found to increase substantially with exercise and faster HRs.¹⁴ A relationship between STD and HR has previously been demonstrated in dogs, with the S-T segment becoming increasingly depressed with shorter R-R intervals.⁸ In humans, an exercise test with the evaluation of an S-T segment/HR slope may be performed to document the presence or absence of STD.¹⁸ The ECGs in this study were all performed at rest, and the population had a median HR of 107 BPM on the ECG. It is possible that reevaluation of the ECG after a period of vigorous exercise would be more valuable.

Although a relationship has been suggested between the presence of myocardial ischemia, STD, and VPCs in both dogs and humans with aortic stenosis, STDV did not correlate with the number of VPCs per 24 hours in this study.³⁴¹⁹ The inability to demonstrate a correlation between STDV and number of VPCs may be due to an inappropriate method of assessing STD as discussed above, or it may suggest that the development of ventricular tachyarrhythmias in these patients is not specifically related to ischemia but may be related to other morphological changes associated with SAS, including myocardial hypertrophy or fibrosis. In a study of humans with SAS, the occurrence of ventricular arrhythmias was not related to the severity of SAS as quantified by PG, but rather to myocardial performance, and significant arrhythmias were seen as an indication of impaired left ventricular function. The authors suggested that the morphological manifestations of ischemia, such as subendocardial fibrosis, may be related to the occurrence of ventricular arrhythmias in these cases.¹⁹

This study has certain limitations, including its retrospective nature as well as the small number of dogs. However, the results of this study indicate that STD is not necessarily a sensitive indicator of SAS severity or frequent VPCs in resting dogs. The S-T segment depression was specific for patients with a PG of ≥80 mm Hg.