A precordial system consists of a group of unipolar ECG leads positioned at different locations on the thorax to record the cardiac electric potential on the horizontal plane.1 Precordial ECG leads are complementary to limb ECG leads and are used in human medicine to properly recognize intraventricular conduction disorders, atrial depolarization, ventricular enlargement, and myocardial infarction and to elaborate several diagnostic algorithms to differentiate wide QRS-complex tachycardia.1,2 The internationally accepted precordial system used in human medicine is the Wilson precordial system, in which 6 unipolar ECG leads are positioned at different locations around the fourth and fifth ICSs. Briefly, precordial leads V1 and V2 are positioned at the right and left sternal margins, respectively; lead V3 is positioned midway between leads V2 and V4; lead V4 is positioned at the left midclavicular line; lead V5 is positioned at the left anterior axillary line; and lead V6 is positioned at the left midaxillary line.2 In humans, atrial activation results in a vector that proceeds in a leftward, superior-inferior, and posterior-anterior direction. Therefore, the P wave is normally upright in leads V3 to V6 and frequently biphasic in V1 and sometimes in V2. Ventricular depolarization consists of 3 vectors. The first vector is derived from septal depolarization and results in a positive deflection in lead V1 and a negative deflection in leads V5 and V6. The second and the third vectors result in a deep negative deflection in lead V1 and positive deflections in leads V2 to V6. Thus, in humans, the standard QRS pattern obtained by use of the Wilson precordial system is characterized by a small R wave and deep S wave (eg, R/S < 1) in V1 and an R wave that becomes larger and an S wave that becomes smaller in the left precordial leads (V2 to V6).3
Over the years, several precordial systems have been proposed for use in veterinary medicine. Some were adapted from human medicine, and others were designed specifically for dogs. In those systems, the lead that monitors ventricular depolarization from the right side of the thorax is positioned in the third or fifth ICS. In 1964, Takahashi4 first proposed the use of a 12-lead precordial system in dogs after correlating electrical potentials at the epicardial surface of the heart with an ECG pattern derived from a lead located on the body surface of mixed-breed dogs.4 For the dogs of that study,4 a precordial lead (C6) located at the CCJ of the right third ICS had a consistent QRS complex of rS type and a positive P wave. In 1965, Detweiler and Patterson5 proposed a modification of the Lannek system as another precordial system for assessment of cardiac function in dogs. That 3-lead system was used in multiple studies,6,7 but it yielded an inconsistent QRS pattern of CV5RL or rV2 (located at the edge of the sternum of the right fifth ICS) with a dominant R wave and smaller S wave. In 2002, Kraus et al8 modified the Wilson precordial system used in humans for dogs and proposed a new 12-lead system, with lead V1 positioned adjacent to the sternum in the right fifth ICS. Although that system is widely used, in our experience, it does not provide a consistent ECG pattern for right atrial and ventricular depolarizations in most dogs, which we propose is because of variation in thoracic conformation among dog breeds. Comparison of thoracic radiographs indicates that the cardiac shape varies among dogs with differing thoracic conformations owing to substantial variation in the dorsoventral diameter of the thorax.9 That variation alters the axis of the heart in the horizontal plane and can influence the ECG pattern obtained by the precordial leads. Therefore, we believe that positioning of the precordial leads in dogs should be adapted on the basis of thoracic conformation. The primary purpose of the study reported here was to determine the accuracy of the Wilson precordial lead system as modified by Kraus et al8 for recording right atrial and ventricular potentials in dogs with different thoracic morphotypes. A secondary goal was to assess alternate positioning of the right precordial lead (V1) in dogs with various thoracic morphotypes to facilitate visualization of right atrial and ventricular depolarization.
Materials and Methods
Animals
The study was approved by the ethical board of the Clinica Veterinaria Malpensa in Samarate, Italy. Sixty healthy client-owned dogs examined at the hospital between June and December 2015 were enrolled in the study following the acquisition of owner consent. Twenty dogs for each of 3 morphotype groups (brachymorphic, mesomorphic, and dolichomorphic) were enrolled in the study. Only dogs without a history of cardiac or respiratory tract disease were eligible for study inclusion. Each dog enrolled in the study also had to be free of abnormalities as determined on the basis of results of a physical examination and cardiovascular evaluation that included 2-view thoracic radiographs, a 5-minute standard 12-lead surface ECG, and standard echocardiographic examination.
At study enrollment, each dog underwent a morphometric study and was allocated to 1 of 3 morphotype groups on the basis of thoracic conformation. With the dog in a standing position, a thorax caliper was used to measure the depth (maximum distance from the dorsum to sternum) and width (maximum distance from the left to right side of the thoracic cavity) of the thorax as described.10 The thoracic index (thoracic width × 100/thoracic depth) was then estimated. Dogs with a thoracic index of 90 to 100, 60 to 89, and 50 to 59 were assigned to the brachymorphic, mesomorphic, and dolichomorphic groups, respectively.11
A digital radiography systema was used to obtain right lateral and dorsoventral thoracic images for each dog. All radiographic images were reviewed and examined by the same investigator (RAS). Cardiac size was evaluated by calculation of the VHS on lateral radiographs as described.12 In general, the reference range for VHS was 9.7 ± 0.5 vertebrae12; however, breed-specific VHS reference ranges were used when that information was available.13–16
Each dog underwent a complete standard transthoracic echocardiographic examination in right and left lateral recumbency. Standard views were obtained by use of 2-D, M-mode, and spectral and color-flow Doppler echocardiographyb with either a 4- or 6-MHz transducer. All echocardiographic examinations and measurements were performed by the same investigator (RAS).
ECG
Each dog underwent a standard 12-lead surface ECGc (standard ECG) while unsedated and positioned in right lateral recumbency. Dogs were instrumented with bipolar limb leads (I, II, and II) and unipolar augmented limb leads (aVR, aVL, and aVF) as described in 1 report5 and a standard precordial lead system as described in another study.8 Briefly, precordial lead V1 was positioned adjacent to the sternum in the fifth ICS. The sixth ICS was used for all left-sided leads (V2 through V6). Lead V2 was positioned adjacent to the sternum, V3 was positioned midway between V2 and V4, V4 was positioned at the CCJ, and V5 and V6 were sequentially positioned dorsal to V4 at a distance equal to that between V3 and V4 (Figure 1). The ECG tracings for all leads were simultaneously recorded for 5 consecutive minutes.
Schematic images of a transverse section of the thorax at the level of the fifth ICS for dogs with brachymorphic, mesomorphic, and dolichomorphic thoracic conformations that also depict the placement of precordial leads as described by Kraus et al8 for a standard 12-lead surface ECG. Lead V1 was positioned adjacent to the sternum in the right fifth ICS. The sixth ICS was used for all left-sided leads (V2 through V6). Lead V2 was positioned adjacent to the sternum, V3 was positioned midway between V2 and V4, V4 was positioned at the CCJ, and V5 and V6 were sequentially positioned dorsal to V4 at a distance equal to that between V3 and V4. The ECG tracings for all leads were simultaneously recorded for 5 consecutive minutes. Notice that the anatomic orientation of the heart and position of V1 relative to the heart varies among the morphotypes; thus, V1 reads the right ventricle in brachymorphic dogs, interventricular septum in mesomorphic dogs, and generally the left ventricle in dolichomorphic dogs.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.358
Following completion of the standard ECG examination, the process was repeated with lead V1 positioned at each of 4 alternate locations on the thoracic wall (CCJ of the right first ICS [1st-R], CCJ of the right third ICS [3rd-R], right third ICS at the level where the thorax was the widest [3rd-H], and CCJ of the left first ICS [1st-L]). Each V1 position evaluated resulted in an R/S < 1 in most dogs regardless of thoracic conformation, which was considered optimal.
All ECG recordings were evaluated by a trained electrocardiographer (DMPV). The QRS complex amplitude, duration, morphology, and electrical axis were electronically assessed for each of the 6 standard limb leads. The mean electrical axis was determined in the frontal plane as follows17: MEA = arctan (Iamp, aVFamp) × 180/π, where MEA is the mean electrical axis, and Iamp and aVFamp are the algebraic sums of the Q, R, and S waves for leads I and aVF, respectively. Standard ECG variables (heart rate, P-wave amplitude and duration, PQ interval duration, QT interval duration, and ST deviation) were evaluated for lead II. For each precordial lead, the ECG parameters evaluated included the QRS complex amplitude and duration, R/S, R-wave amplitude progression (left precordial leads [V2 to V6] only), and P-wave amplitude and duration. For each lead, each ECG variable was measured for 3 randomly selected heart beats, and the mean was calculated and used for analysis purposes. Previously described18 reference ranges were used to assess each ECG variable.
Statistical analysis
Commercially available softwared was used to perform all statistical analyses. The Shapiro-Wilk test was used to assess the data distribution of each continuous variable for normality. The mean ± SD was used to summarize results for normally distributed continuous variables, and the median (IQR) was used to summarize results for nonnormally distributed continuous variables. Results for categorical variables were summarized as the number and percentage of dogs. A χ2 test was used to evaluate whether sex was associated with thoracic morphotype. A Kruskal-Wallis test was used to evaluate whether age was associated with thoracic morphotype. Because none of the ECG variables were normally distributed, the Kruskal-Wallis test was also used to compare each of those variables among the 3 thoracic morphotypes (brachymorphic, mesomorphic, and dolichomorphic). When pairwise comparisons were necessary (ie, the Kruskal-Wallis test yielded a value of P < 0.05), post hoc analyses were performed by means of the Dunn test with the Bonferroni method used to adjust P values and avoid type I error inflation. The respective associations between morphotype and the P-wave polarity of lead V1 (positive or negative) and R/S (< or ≥ 1) were assessed with χ2 tests. The data for each ECG variable for V1 were rank transformed to allow for parametric analysis. A 2-way ANOVA for repeated measures (ie, mixed generalized linear model) was used to assess the effect of V1 location (5th [control], 1st-L, 1st-R, 3rd-R, or 3rd-H) and thoracic morphotype on each ECG variable (dependent variable). Each model included fixed effects for the thoracic morphotype, V1 location, and interaction between thoracic morphotype and V1 location and a random effect for study subject to account for repeated measures within dogs. The Bonferroni method was used when post hoc analyses were necessary.
For each ECG variable, the intrarater reliability was assessed by calculation of the ICC for the 3 measurements obtained from 3 randomly selected heart beats. An ICC < 0.5 was considered low reliability, between 0.5 and 0.8 was considered moderate reliability, and > 0.8 was considered high reliability.
Results
Dogs
The brachymorphic group consisted of 10 English Bulldogs, 6 French Bulldogs, and 4 Pugs. There were 7 sexually intact females, 6 spayed females, and 7 sexually intact males. The group had a mean ± SD age of 4.8 ± 3.7 years and body weight of 17.8 ± 8.6 kg and a median BCS of 8 (IQR, 1.25) on a scale of 1 to 9. The median thoracic index was 95.5 (IQR, 22.36) for the brachymorphic group.
The mesomorphic group consisted of 7 Golden Retrievers, 3 mixed-breed dogs, 2 Australian Shepherds, 2 Boxers, 2 Segugio Italianos, 1 German Shepherd Dog, 1 Cane Corso, 1 Jack Russell Terrier, and 1 Labrador Retriever. There were 8 sexually intact females, 2 spayed females, 6 sexually intact males, and 4 neutered males. The group had a mean ± SD age of 4.7 ± 2.7 years and body weight of 26.4 ± 9.3 kg and a median BCS of 7 (IQR, 1.25). The median thoracic index was 73.6 (IQR, 5.89) for the mesomorphic group.
The dolichomorphic group consisted of 12 Greyhounds, 5 Doberman Pinschers, 1 Afghan Hound, 1 Borzoi, and 1 Ibizan Hound. There were 5 sexually intact females, 2 spayed females, 2 sexually intact males, and 11 neutered males. The group had a mean ± SD age of 4.9 ± 1.8 years and body weight of 34.5 ± 6.0 kg and a median BCS of 7 (IQR, 1.0). The median thoracic index was 57.7 (IQR, 5.49) for the dolichomorphic group. The sex distribution (P = 0.20) and mean age (P = 0.65) did not differ significantly among the 3 morphotype groups.
ECG variables
The ICC ranged from 0.84 to 0.99 for all ECG variables evaluated, which indicated that the intrarater reliability was high for all measurements. Data were not normally distributed for any of the standard ECG variables except P-wave duration. Electrocardiographic variables (Table 1) and QRS morphological measurements (Table 2) for the bipolar limb leads (I, II, and III) and unipolar augmented limb leads (aVR, aVL, and aVF) determined during the standard ECG examination were summarized.
Median (IQR) for ECG variables determined from lead II during a standard 12-lead surface ECG for healthy adult dogs with a brachymorphic (n = 20), mesomorphic (20), or dolichomorphic (20) thoracic conformation.
| Variable | Brachymorphic group | Mesomorphic group | Dolichomorphic group |
|---|---|---|---|
| Heart rate (beats/min) | 137.5 (23) | 114 (39)* | 131 (26) |
| P-wave duration (ms) | 42 (5.75) | 44.5 (9.25) | 42 (8.75) |
| P-wave amplitude (mV) | 0.12 (0.045) | 0.15 (0.063) | 0.22 (0.135)* |
| QRS duration (ms) | 55 (5.5) | 55 (6.5) | 56.5 (8.3) |
| QRS electrical axis (°) | 57.48 (24.81) | 83.85 (9.67) | 80.09 (32.26) |
| PQ interval (ms) | 95 (14.25) | 107.6 (28.5)* | 107 (27.3)* |
| QT interval (ms) | 187 (28.3) | 189 (38.5) | 82 (19.0) |
The median thoracic index (thoracic width × 100/thoracic depth) was 95.5 (IQR, 22.36) for the brachymorphic group, 73.6 (IQR, 5.89) for the mesomorphic group, and 57.7 (IQR, 5.49) for the dolichomorphic group. For each variable, a Kruskal-Wallis test was used for global comparison among the 3 thoracic conformation groups; when necessary, pairwise comparisons were performed by means of the Dunn test with a Bonferroni correction to control for type I error inflation.
Value differs significantly (P < 0.05) from that for the brachymorphic group.
Median (IQR) amplitude (mV) of the Q, R, and S waves for the bipolar limb leads (I, II, and III) and unipolar augmented limb leads (aVR, aVL, and aVF) during a standard 12-lead surface ECG for the dogs of Table 1.
| Brachymorphic group | Mesomorphic group | Dolichomorphic group | |||||
|---|---|---|---|---|---|---|---|
| Lead | Wave | Median (IQR) | No. of dogs with a value other than 0 | Median (IQR) | No. of dogs with a value other than 0 | Median (IQR) | No. of dogs with a value other than 0 |
| I | Q | 0.03 (0.0525) | 12 | 0.23 (0.1225)* | 19 | 0 (0.025)* | 6 |
| R | 0.22 (0.1525) | 20 | 0.23 (0.16) | 20 | 0.255 (0.207) | 20 | |
| S | 0 (0.025) | 5 | 0 (0) | 0 | 0 (0) | 4 | |
| II | Q | 0.01 (0.043) | 10 | 0.07 (0.07)* | 19 | 0.11 (0.09)* | 19 |
| R | 0.44 (0.19) | 20 | 1.02 (0.87)* | 20 | 1.42 (1.045)* | 20 | |
| S | 0.05 (0.063) | 12 | 0.03 (0.060) | 20 | 0.04 (0.062) | 14 | |
| III | Q | 0 (0.0225) | 6 | 0.07 (0.0900)* | 15 | 1.11 (1.01)*† | 19 |
| R | 0.30 (0.1925) | 20 | 0.90 (0.6400)* | 20 | 1.26 (1.305)* | 20 | |
| S | 0.04 (0.0450) | 17 | 0.02 (0.0800) | 13 | 0.03 (0.052) | 13 | |
| aVR | Q | 0 (0.2775) | 9 | 0 (0.0500) | 2 | 0 (0) | 3 |
| R | 0.03 (0.0250) | 20 | 0.08 (0.7500)* | 20 | 0.07 (0.045)* | 20 | |
| S | 0.14 (0.3850) | 11 | 0.66 (0.0700)* | 18 | 0.75 (0.463)* | 17 | |
| aVL | Q | 0.105 (0.1225) | 17 | 0.145 (0.3650)* | 11 | 0 (0)*† | 4 |
| R | 0.045 (0.0725) | 20 | 0.045 (0.0700) | 20 | 0.08 (0.0575)* | 20 | |
| S | 0.0 (0.000) | 3 | 0.0 (0.5300) | 10 | 0.28 (0.7)* | 13 | |
| aVF | Q | 0 (0.0225) | 8 | 0.05 (0.000)* | 18 | 0.12 (0.065)*† | 20 |
| R | 0.34 (0.2050) | 20 | 0.89 (0.0600)* | 20 | 1.33 (1.247)* | 20 | |
| S | 0.04 (0.0450) | 17 | 0.04 (0.5000) | 14 | 0.04 (0.725) | 14 | |
For each lead, the number of dogs with a value other than 0 (ie, the number of dogs that contributed to each median [IQR]) is also provided.
Value differs significantly (P < 0.05) from that for the mesomorphic group.
See Table 1 for remainder of key.
The ECG morphological measurements for the unipolar precordial limb leads (V1, V2, V3, V4, V5, and V6) during the standard ECG examination were likewise summarized (Table 3). For all dogs, the P wave had positive polarity in all precordial leads during the standard ECG examination. The Q wave was typically absent for leads V1, V2, V3, and V4 and was variably present for leads V5 and V6. The Q wave was present in lead V5 for 7, 7, and 13 dogs in the brachymorphic, mesomorphic, and dolichomorphic groups, respectively. The Q wave was present in lead V6 for 10, 10, and 14 dogs in the brachymorphic, mesomorphic, and dolichomorphic groups, respectively. In both leads V5 and V6, the median Q-wave amplitude for the brachymorphic group was significantly less than that for the dolichomorphic group but did not differ significantly from the median Q-wave amplitude for the mesomorphic group. The median Q-wave amplitude did not differ significantly between the mesomorphic and dolichomorphic groups. An R wave was present in all precordial leads for all dogs. The median R-wave amplitude for the brachymorphic group was significantly less than that for both the mesomorphic and dolichomorphic groups in all leads. For all 3 morphotype groups, the median R-wave amplitude progressively increased from leads V1 to V4 and then progressively decreased from leads V4 to V6 (Figure 2). Most dogs had an S wave present in all precordial leads. The median S-wave amplitude did not differ significantly among the 3 groups for any precordial lead except lead V6. For lead V6, the median S-wave amplitude for the brachymorphic group was significantly less than that for the dolichomorphic group. All dogs had an R/S ≥ 1 for all precordial leads except V1. For lead V1, the R/S was < 1 for 16, 6, and 7 dogs of the brachymorphic, mesomorphic, and dolichomorphic groups, respectively, and the proportion of dogs with an R/S < 1 for the dolichomorphic group was significantly (P = 0.002) lower than that for the brachymorphic group. The median R/S varied significantly among the 3 morphotype groups for all precordial leads except V6. The median R/S was highest for the mesomorphic group for leads V1 through V5.
Box-and-whisker plots of the amplitude of the Q (A, D, and G), R (B, E, and H), and S (C, F, and I) waves in the precordial leads (V1 through V6) located as described in Figure 1 for healthy dogs with a brachymorphic (A, B, and C; n = 20), mesomorphic (D, E, and F; 20), and dolichomorphic (G, H, and I; 20) thoracic conformation. For each plot, the upper and lower limits of the box delimit the IQR (25th and 75th percentiles), the horizontal line within the box represents the median, and the whiskers delimit the range. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.358
Median (IQR) for ECG variables for unipolar precordial leads during a standard 12-lead surface ECG for the dogs of Table 1.
| Lead | Variable | Brachymorphic group | Mesomorphic group | Dolichomorphic group |
|---|---|---|---|---|
| V1 | P-wave duration (ms) | 35.0 (7.75) | 46.5 (14.75)* | 50 (17.50)* |
| P-wave amplitude (mV) | 0.02 (0.02) | 0.05 (0.05) | 0.09 (0.06)*† | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| R-wave amplitude (mV) | 0.19 (0.18) | 0.79 (0.64)* | 0.79 (0.32)* | |
| S-wave amplitude (mV) | 0.45 (0.27) | 0.41 (0.35) | 0.54 (0.69) | |
| R/S | 0.46 (0.58) | 1.54 (4.95)* | 1.15 (2.67)* | |
| V2 | P-wave duration (ms) | 45.5 (9.75) | 54.0 (6.50)* | 52.0 (12.75)* |
| P-wave amplitude (mV) | 0.08 (0.05) | 0.11 (0.05)* | 0.15 (0.10)*† | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.01) | 0 (0.00) | |
| R-wave amplitude (mV) | 0.51 (0.31) | 1.44 (0.53)* | 1.68 (0.83)* | |
| S-wave amplitude (mV) | 0.17 (0.16) | 0.14 (0.11) | 0.24 (0.23) | |
| R/S | 2.26 (2.84) | 11.83 (12.94)* | 7.65 (6.79) | |
| V3 | P-wave duration (ms) | 46.0 (8.50) | 56.0 (6.25)* | 51.0 (7.25)*† |
| P-wave amplitude (mV) | 0.08 (0.04) | 0.13 (0.05)* | 0.17 (0.13)*† | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.01) | 0 (0.01) | |
| R-wave amplitude (mV) | 0.55 (0.36) | 1.41 (0.77)* | 1.75 (0.90)* | |
| S-wave amplitude (mV) | 0.18 (0.15) | 0.11 (0.11) | 0.27 (0.24) | |
| R/S | 2.60 (3.76) | 17.71 (15.04)* | 8.21 (7.58)* | |
| V4 | P-wave duration (ms) | 47.50 (7.25) | 53.00 (9.50)* | 51.00 (8.75) |
| P-wave amplitude (mV) | 0.10 (0.05) | 0.14 (0.02) | 0.20 (0.15)*† | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.03) | 0 (0.07) | |
| R-wave amplitude (mV) | 0.57 (0.27) | 1.42 (0.76)* | 0.76 (0.93)* | |
| S-wave amplitude (mV) | 0.11 (0.08) | 0.09 (0.09) | 0.19 (0.33) | |
| R/S | 4.78 (4.20) | 13.30 (15.12)* | 8.44 (9.84) | |
| V5 | P-wave duration (ms) | 44.50 (10.25) | 51.50 (6.50)* | 52.00 (7.25)* |
| P-wave amplitude (mV) | 0.09 (0.05) | 0.15 (0.05) | 0.22 (0.16)*† | |
| Q-wave amplitude (mV) | 0 (0.02) | 0 (0.04) | 0.04 (0.07)* | |
| R-wave amplitude (mV) | 0.5 (0.36) | 1.36 (0.92)* | 1.46 (0.77)* | |
| S-wave amplitude (mV) | 0.07 (0.10) | 0.08 (0.06) | 0.13 (0.23) | |
| R/S | 4.7 (7.79) | 16.51 (13.74)* | 8.33 (11.97) | |
| V6 | P-wave duration (ms) | 42.50 (10.50) | 50.00 (7.50)* | 48.50 (5.50) |
| P-wave amplitude (mV) | 0.09 (0.03) | 0.13 (0.04)* | 0.21 (0.19)*† | |
| Q-wave amplitude (mV) | 0.005 (0.02) | 0.005 (0.06) | 0.05 (0.09)* | |
| R-wave amplitude (mV) | 0.49 (0.30) | 1.13 (0.65)* | 1.21 (0.66)* | |
| S-wave amplitude (mV) | 0.06 (0.07) | 0.05 (0.06) | 0.10 (0.10)* | |
| R/S | 6.28 (9.71) | 10.28 (19.38) | 17.8 (11.14) |
Lead V1 was positioned adjacent to the sternum in the right fifth ICS. The sixth ICS was used for all left-sided leads (V2 through V6). Lead V2 was positioned adjacent to the sternum, V3 was positioned midway between V2 and V4, V4 was positioned at the CCJ, and V5 and V6 were sequentially positioned dorsal to V4 at a distance equal to that between V3 and V4.
The ECG morphological measurements for lead V1 at each of the 4 alternate locations assessed were summarized (Table 4). For many dogs, the P wave had negative polarity at all 4 alternate locations; however, the largest proportion of dogs had a negative P wave when V1 was located at the 1st-R position (Table 5). The median R/S was < 1 for all 3 morphotype groups and did not differ significantly among the 3 groups at any of the alternate locations except 3rd-R. When V1 was at the 3rd-R position, the median R/S for the mesomorphic group (1.1) was significantly greater than that for the brachymorphic group (0.35) but did not differ significantly from the median R/S for the dolichomorphic group (0.62). The R wave was absent (QS complex) in 8 and 5 dogs of the brachymorphic group when lead V1 was at the 3rd-H and 1st-L locations, respectively. Two dogs of the dolichomorphic group had QS complexes when V1 was at the 3rd-H location. When the R/S data distribution for V1 was compared among the 5 positions assessed (1st-R, 3rd-R, 3rd-H, 1st-L, and 5th [adjacent to the sternum in the right fifth ICS; control]), it was most consistent (ie, had the least variation) when it was at the 1st-R location (Figure 3). Moreover, the median R/S for V1 at the 1st-R position was significantly lower than the median R/S for V1 at the control position for all 3 morphotype groups, and that was the only position evaluated where the median R/S for V1 differed significantly from that for the control. Schematic images of the location of V1 in the 1st-R position relative to the heart and other precordial leads for each thoracic morphotype were created (Figure 4).
Box-and-whisker plots of the R/S in lead V1 when it was located adjacent to the sternum in the right fifth ICS (5th; control), at the CCJ of the right first ICS (1st-R), at the CCJ of the right third ICS (3rd-R), at the right third ICS where the thorax was the widest (3rd-H), and at the CCJ of the left first ICS (1st-L) during 12-lead surface ECG for dogs with brachymorphic (A), mesomorphic (B), and dolichomorphic (C) thoracic conformations. Circles represent outliers. See Figure 2 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.358
Schematic images of the lateral aspect of the thorax for dogs with brachymorphic, mesomorphic, and dolichomorphic thoracic conformations that depict placement of the precordial leads when V1 was located at the CCJ of the right first ICS (I; 1st-R location). The sixth ICS (VI) was used for all left-sided leads (V2 through V6), and those leads were located as described in Figure 1. Notice that the orientation of the heart within the thorax and the location of lead V1 in relation to the heart varies among the 3 morphotypes. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.358
Median (IQR) for ECG variables for precordial lead V1 when it was positioned at each of 4 alternate locations on the thorax of the dogs of Table 3.
| V1 position | Variable | Brachymorphic group | Mesomorphic group | Dolichomorphic group |
|---|---|---|---|---|
| 1st-R | P-wave duration (ms) | 31.5 (4.50) | 36 (5.25) | 32.5 (12.25) |
| P-wave amplitude (mV) | −0.03 (0.02) | −0.03 (0.06) | −0.03 (0.07) | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| R-wave amplitude (mV) | 0.08 (0.10) | 0.2 (0.17)* | 0.25 (0.21)* | |
| S-wave amplitude (mV) | 0.3 (0.13) | 0.48 (0.35) | 0.71 (0.45)*† | |
| R/S | 0.24 (0.34) | 0.57 (0.41) | 0.28 (0.70) | |
| 3rd-R | P-wave duration (ms) | 37 (8.25) | 40 (8.50)* | 34 (8.00)* |
| P-wave amplitude (mV) | −0.04 (0.03) | 0.04 (0.08)* | 0.04 (0.05)* | |
| Q-wave amplitude (mV) | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| R-wave amplitude (mV) | 0.14 (0.10) | 0.52 (0.35)* | 0.69 (0.26)*† | |
| S-wave amplitude (mV) | 0.33 (0.24) | 0.60 (0.37) | 0.89 (1.32)*† | |
| R/S | 0.35 (0.53) | 1.1 (1.17)* | 0.62 (1.24) | |
| 3rd-H | P-wave duration (ms) | 37.5 (4.25 | 37.5 (8.50) | 39 (11.25) |
| P-wave amplitude (mV) | −0.05 (0.02) | −0.04 (0.10)* | −0.09 (0.14)† | |
| Q-wave amplitude (mV) | 0 (0.13) | 0 (0.00) | 0 (0.01)† | |
| R-wave amplitude (mV) | 0.07 (0.09) | 0.32 (0.38)* | 0.29 (0.49)* | |
| S-wave amplitude (mV) | 0.11 (0.30) | 0.37 (0.54)* | 0.27 (0.87)* | |
| R/S | 0.31 (0.85) | 0.67 (2.62) | 0.24 (3.02) | |
| 1st-L | P-wave duration (ms) | 31.5 (9.50) | 37.5 (6.50) | 36 (9.00) |
| P-wave amplitude (mV) | 0.01 (0.05) | −0.02 (0.05) | 0 (0.07) | |
| Q-wave amplitude (mV) | 0 (0.02) | 0 (0.00) | 0 (0.00) | |
| R-wave amplitude (mV) | 0.06 (0.13) | 0.12 (0.09) | 0.17 (0.26)* | |
| S-wave amplitude (mV) | 0.05 (0.18) | 0.15 (0.24) | 0.33 (0.37)* | |
| R/S | 0.83 (2.52) | 0.36 (3.33) | 0.39 (0.80) |
Number of dogs within each category of select ECG variables for precordial lead V1 when it was positioned at each of 4 alternate locations on the thorax of the dogs of Table 1.
| V1 position | Variable | Category | Brachymorphic group | Mesomorphic group | Dolichomorphic group | P value* |
|---|---|---|---|---|---|---|
| 1st-R | P-wave polarity | Negative | 20 | 15 | 14 | 0.91 |
| Positive | 0 | 5 | 6 | |||
| R/S | < 1 | 19 | 17 | 16 | 0.08 | |
| ≥ 1 | 1 | 3 | 4 | |||
| QS | 0 | 0 | 0 | |||
| 3rd-R | P-wave polarity | Negative | 16 | 8 | 4 | < 0.001 |
| Positive | 4 | 12 | 16 | |||
| R/S | < 1 | 14 | 9 | 12 | 0.56 | |
| ≥ 1 | 6 | 11 | 8 | |||
| QS | 0 | 0 | 0 | |||
| 3rd-H | P-wave polarity | Negative | 20 | 14 | 14 | 0.01 |
| Positive | 0 | 6 | 6 | |||
| R/S | < 1 | 9 | 10 | 9 | 0.10 | |
| ≥ 1 | 3 | 10 | 9 | |||
| QS | 8 | 0 | 2 | |||
| 1st-L | P-wave polarity | Negative | 9 | 11 | 10 | 0.81 |
| Positive | 11 | 9 | 10 | |||
| R/S | < 1 | 8 | 10 | 14 | 0.94 | |
| ≥1 | 7 | 10 | 6 | |||
| QS | 5 | 0 | 0 |
Results of mixed generalized linear modeling indicated that both P-wave duration and P-wave amplitude for lead V1 were significantly associated with V1 location (P < 0.001), thoracic morphotype (P < 0.001), and the interaction between V1 location and thoracic morphotype (P < 0.001). Moreover, the P-wave duration and amplitude both varied in the same manner. The median P-wave duration and amplitude for each of the 4 alternate V1 locations differed significantly from the median P-wave duration and amplitude for the control location. Regardless of V1 location, the median P-wave duration and amplitude for the brachymorphic group were significantly (P = 0.001) less than the median P-wave duration and amplitude for the mesomorphic group. Within the brachymorphic group, the median P-wave duration and amplitude for the 3rd-R and 3rd-H locations differed significantly from those for the control location. Within the mesomorphic group, the median P-wave duration and amplitude for the 1st-R and 1st-L locations differed significantly from those for the control location. Within the dolichomorphic group, the median P-wave duration and amplitude for the 1st-R, 1st-L, and 3rd-H locations differed significantly from those for the control location (Tables 3 and 4).
Discussion
During an ECG examination, the precordial system complements the limb system by analyzing the cardiac electrical potential within the horizontal plane. Results of a previous study4 involving dogs indicate that the electrical potential recorded at the right atrial epicardium corresponds to a negative P wave on the surface ECG, and the electrical potential recorded at the right ventricular epicardium corresponds to a small R wave and a deep S wave. For the mesomorphic mongrel dogs of that study,4 the precordial system used reproduced the correct right atrial and right ventricular electrical potentials on the surface ECG. Precordial leads C3 (left fifth ICS at the CCJ), C4 (right seventh ICS at the CCJ), M2 (left sixth ICS at the widest portion of the thorax), M3 (just left of the xiphoid process of the sternum), M4 (just right of the xiphoid process of the sternum), and M5 (right seventh ICS at the widest portion of the thorax) had an Rs pattern derived from the left ventricle and interventricular septum and a positive P wave.4 Precordial leads C1 (cranial to the left first rib at the CCJ) and M1 (left third ICS at the widest portion of the thorax) had Qr and QR patterns derived from the interventricular septum, the posterior wall of the left ventricle, and part of the right ventricle.4 The P wave was negative in C1 and positive in M1.4 An intermediate RS pattern representative of the anterior wall of the left ventricle, interventricular septum, and part of the right ventricle was evident in leads C2 and C5 when those leads were positioned at the CCJ of the left second ICS and right fifth ICS, respectively.4 The P wave was positive or negative in C2 and positive in C5.4 Leads C6 and M6 had an rS pattern, corresponding to right ventricular depolarization when the leads were positioned in the right third ICS at the CCJ and widest aspect of the thorax, respectively.4 For the dogs of the present study during the standard 12-lead ECG, lead V1 was positioned adjacent to the sternum in the right fifth ICS as described by Kraus,8 and it did not have an rS pattern with an R/S < 1 in most dogs. When lead V1 was placed at the CCJ of the right first ICS (1st-R location), it had a P wave with a negative component and a fairly consistent QRS pattern with an R/S < 1 in most dogs regardless of thoracic morphotype in a manner similar to that observed for V1 in humans.3 The vector of excitation from the sinus node to the atrioventricular node is directed leftward in a posterior-anterior and superior-inferior direction, whereas the vector associated with depolarization of the right ventricle is directed rightward in a posterior-anterior and inferior-superior direction. Consequently, the P wave is typically upright in leads V3 to V6 and biphasic in lead V1 (and sometimes in lead V2).3 Additionally, the QRS complex in lead V1 has an rS morphology with an R/S < 1.3
In our clinical experience, many precordial systems proposed for use in dogs do not provide consistent evaluation of right ventricular depolarization. Furthermore, prior to the present study, none of the precordial systems described for dogs took into consideration the various thoracic conformations and position of the heart within the thorax of different breeds. In fact, most of those systems placed lead V1 close to the sternum in the right fifth ICS, regardless of thoracic morphotype. In the present study, when lead V1 was placed adjacent to the sternum in the right fifth ICS (5th [control] location) as previously described,8 it provided consistent evaluation of right ventricular depolarization (ie, R/S < 1) for only 16 of 20 dogs in the brachymorphic group, 6 of 20 dogs in the mesomorphic group, and 7 of 20 dogs in the dolichomorphic group. Relative to the control location, when lead V1 was placed at the 1st-R location, it resulted in a smaller R wave rather than a taller S wave in brachymorphic dogs, which indicated that the exploring electrode was reading from the right ventricle with the first septal vector approaching and the third left ventricular vector moving away.
In human medicine, the primary factors that affect the appearance (eg, axis and amplitude) of the QRS complex on the surface ECG are the position of the heart within the thorax and patient age, sex, race, body mass, and physical fitness.19 In regard to the electrical axis of the QRS complex, the heart has 3 main positions (horizontal, intermediate, and vertical) in the frontal plane. The heart is in a vertical position if the QRS axis is oriented at ≥ 60° and in a horizontal position if the QRS axis is oriented at > 30° but < 60°.19 In veterinary medicine, the cardiac silhouette on thoracic radiographs varies among species and among breeds of the same species, primarily owing to the position of the heart within the thoracic cavity.9 Given basic ECG principles and descriptions by Detweiler20 and Gonin,e the electrical axis of the QRS complex in the frontal plane varies with body conformation. Narrow-chested dogs have a fairly constant vertical electrical axis, compared with broad-chested dogs, which have a more horizontal and more variable electrical axis. The median electrical axis of the heart in the frontal plane was 83.85° and 80.09° for dogs in the mesomorphic and dolichomorphic groups, respectively. Those dogs typically had a Q wave present in the inferior (II, III, and aVF) leads, but the Q wave was only occasionally present in the superior leads, which suggested that the heart axis was shifted vertically and cranially, compared with that for brachymorphic dogs. The median electrical axis of the heart in the frontal plane was 57.48° for brachymorphic dogs, and the Q wave was more frequently present in lead aVL than in leads III and aVF, which suggested that the heart was oriented along a fairly horizontal axis. The position of the heart within the thorax affects the resultant vector recorded by lead V1. Therefore, if the heart is vertically positioned within the thorax, such as in mesomorphic and dolichomorphic dogs, it is probable that placement of lead V1 at the control location will yield a QRS pattern from the interventricular septum or left ventricle. However, we assume that when such dogs are positioned in right lateral recumbency, the heart will shift to the right, which will cause the interventricular septum to remain close to the right side of the sternum near V1.
In the present study, the dogs in all 3 thoracic morphotype groups had a QRS complex that was characterized by a predominant R wave (ie, R/S > 1) in the left precordial leads (V2 through V6), the amplitude of which progressively increased from leads V2 to V5 and then decreased from leads V5 to V6. Those findings confirmed that the left precordial leads were correctly positioned and reflected normal ventricular depolarization. Alterations in the R-wave amplitude, particularly with a decrease in amplitude between leads V5 and V6, can be the result of an attenuating effect of the lungs or by the more dorsal position of V6 on the thorax relative to V5. A similar phenomenon has been described in human medicine.21
Among the dogs of the present study, the R-wave amplitude in the limb and precordial leads was lowest for the brachymorphic group. The electrical potential voltage recorded by the precordial leads is dependent on the direction and magnitude of the resultant cardiac vector generated from the epicardium to the body surface. In human medicine, some cardiac (myocardial infarctions, infiltrative cardiomyopathies, impaired cardiac volume, or pericardial diseases) and extracardiac (an increase in distance between the heart and thoracic wall, abnormally decreased pulmonary impedance, and position of precordial leads) causes can alter the transfer of heart potentials to the body surface and produce a QRS complex with low voltage in the frontal and precordial leads.22 Brachymorphic dogs have a greater heart-to-thoracic wall distance than mesomorphic and dolichomorphic dogs, likely because of their fairly round thoracic shape and the propensity for those dogs to have more body fat than dogs of other morphotypes. Those characteristics, in conjunction with a fairly horizontal electrical axis in the frontal plane, could induce a low QRS amplitude in the surface ECG.
Limitations of the present study included the small number of dogs evaluated for each thoracic morphotype and the small number of lead V1 locations assessed. Dogs were enrolled in the study on the basis of thoracic morphotype, and multiple breeds were represented in each group. Breed-specific thoracic conformation may affect the position of the heart within the thorax, and other breeds of dogs need to be evaluated. Also, we did not assess factors such as age and sex. Only 4 alternate lead V1 locations were assessed. Moreover, the accuracy of electrode placement by the operator was not evaluated, and we cannot exclude variation in electrode positioning among study subjects despite the use of anatomic references.
Results of the present study indicated that placement of lead V1 at the 5th ICS (control) location yielded an inconsistent right ventricular ECG pattern in dogs with various thoracic conformations. When lead V1 was at the control location, it provided consistent evaluation of right atrial and ventricular depolarization for most brachymorphic dogs but not for most mesomorphic or dolichomorphic dogs. Placement of lead V1 at the CCJ of the right first ICS (1st-R location) resulted in the most consistent evaluation of right atrial and ventricular depolarization for dogs of all 3 thoracic morphotypes and should be evaluated further in precordial lead systems for dogs undergoing 12-lead surface ECG.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript.
The authors declare that there were no conflicts of interest.
The authors thank Dr. Nicolletta Vitali for assistance with statistical analysis, Drs. Eva Oxford and Flavia Giacomazzi for assistance with manuscript preparation, and Alfonso Tortora and Federica Faré for artwork.
ABBREVIATIONS
| BCS | Body condition score |
| CCJ | Costochondral junction |
| ICC | Intraclass correlation coefficient |
| ICS | Intercostal space |
| IQR | Interquartile range (75th percentile – 25th percentile) |
| R/S | R wave-to-S wave ratio |
| VHS | Vertebral heart score |
Footnotes
Agfa Health Care, Milan, Italy.
Vivid E9, GE Healthcare, Milan, Italy.
Easy ECG pocket, ATES Medical, Verona, Italy.
SAS, version 9.2, SAS Institute Inc, Cary, NC.
Gonin P. Über die Lage der elektrischen Herzachse beim Hund. Inaugural dissertation, Vetsuisse Faculty, University of Bern, Bern, Switzerland, 1962.
References
1. Wilson FN. The clinical value of chest leads. Br Heart J 1948;10:88–91.
2. Kligfield P, Gettes LS, Bailey JJ, et al. Recommendations for the standardization and interpretation of the electrocardiogram. Part I: the electrocardiogram and its technology: a scientific statement from the American Heart Association Electrocardiography and Arrhythmias Committee, Council on Clinical Cardiology; the American College of Cardiology Foundation; and the Heart Rhythm Society. Heart Rhythm 2007;4:394–412.
3. Goldman MJ. Definitions of electrocardiographic configurations. In: Principles of clinical electrocardiography. 11th ed. Los Altos, Calif: Lange Medical Publications, 1982;23–28.
4. Takahashi M. Experimental studies on the electrocardiogram of the dog. Nippon Juigaku Zasshi 1964;26:191–210.
5. Detweiler DK, Patterson DF. The prevalence and types of cardiovascular disease in dogs. Ann N Y Acad Sci 1965;127:481–516.
6. Detweiler DK. The use of electrocardiography in toxicological studies with Beagle dogs. In: Balazs T, ed. Cardiac toxicology, Vol 3. Boca Raton, Fla: CRC Press, 1981;33–82.
7. Hill JD. The electrocardiogram in dogs with standardized body and limb positions. J Electrocardiol 1968;1:175–182.
8. Kraus MS, Moïse NS, Rishniw M, et al. Morphology of ventricular arrhythmias in the Boxer as measured by 12-lead electrocardiography with pace-mapping comparison. J Vet Intern Med 2002;16:153–158.
9. Coulson A, Lewis N. Thorax. In: An atlas of interpretative radiographic anatomy of the dog and cat. 2nd ed. Oxford, England: Wiley-Blackwell, 2008;311–312.
10. Canton M. Misure. Tronco. In: Prontuario di cinometria. Porto Viro, Italy: Antonio Crepaldi, 2013;125.
11. Bonetti F. Le regioni del tronco. In: Carosio R, ed. Zoognostica del cane. Bologna, Italy: Sestante Edizioni, 1995;303.
12. Buchanan JW, Bücheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc 1995;206:194–199.
13. Bavegems V, Van Caelenberg A, Duchateau L, et al. Vertebral heart size ranges specific for Whippets. Vet Radiol Ultrasound 2005;46:400–403.
14. Jepsen-Grant K, Pollard RE, Johnson LR. Vertebral heart scores in eight dog breeds. Vet Radiol Ultrasound 2013;54:3–8.
15. Steel JD, Taylor RI, Davis PE, et al. Relationships between heart score, heart weight and body weight in Greyhound dogs. Aust Vet J 1976;52:561–564.
16. Lamb CR, Wikeley H, Boswood A, et al. Use of breed-specific ranges for the vertebral heart scale as an aid to the radiographic diagnosis of cardiac disease in dogs. Vet Rec 2001;148:707–711.
17. Rishniw M, Porciello F, Erb HN, et al. Effect of body position on the 6-lead ECG of dogs. J Vet Intern Med 2002;16:69–73.
18. Detweiler DK. The dog electrocardiogram: a critical review. In: MacFarlane PW, van Oosterom A, Pahlm O, et al, eds. Comprehensive electrocardiology. 2nd ed. London: Springer, 2010;1881–1883.
19. Macfarlane PW, Veitch Lawrie TD. The normal electrocardiogram and vectocardiogram. In: MacFarlane PW, van Oosterom A, Pahlm O, et al, eds. Comprehensive electrocardiology. 2nd ed. London: Springer, 2010;503–511.
20. Detweiler DK. Cardiovascular disease in animals, clinical consideration. In: Luisada AA, ed. Cardiology. Vol 5. New York: McGraw-Hill, 1961;10–42.
21. MacKenzie R. Poor R-wave progression. J Insur Med 2005;37:58–62.
22. Madias JE. Low QRS voltage and its causes. J Electrocardiol 2008;41:498–500.
