The quantitative assessment of LV systolic function is pivotal to the diagnosis and treatment of cardiovascular disease. Invasive indices of LV contractility have traditionally been considered the gold standard for such assessment, and 2 such indices are most often used in research settings. The end-systolic pressure volume relationship with calculation of ventricular elastance is probably least influenced by changes in heart rate, preload, and afterload; however, the technique used to determine elastance is technically demanding. The measurement of LV+dp/dtmax is a relatively sensitive estimate of global ventricular contractility in conditions of constant ventricular loading. Although this variable is greatly influenced by preload, it is less affected by changes in afterload.1,2 The relative ease of obtaining LV function estimates with this technique has led to its common use for validation of noninvasive indices of ventricular contractility or global systolic function.3,4
The most clinically applicable index of LV systolic function would be one that is noninvasive as well as sensitive and specific for early detection of pathological changes, repeatable, and easily performed. Echocardiography fulfills most of these requirements; however, most conventional echocardiographic indices of LV systolic function, such as fractional shortening, shortening area, and ejection fraction, are highly load dependent and may be insensitive to regional or early ventricular disease.5,6 Similarly, systolic time intervals from the onset of the QRS complex to the onset of ejection (PEP) and time duration of ejection (ejection time) are affected by ventricular loading.
Myocardial strain is a measure of absolute tissue deformation and is defined as the percentage of change of a segment of myocardium from its original length. Strain rate is the rate of deformation and is the first temporal derivative of strain. By convention, myocardial thickening (radial plane) or lengthening (longitudinal plane) yields a positive strain value, whereas myocardial thinning (radial plane) or shortening (longitudinal plane) yields a negative strain value.7–10 Segmental systolic strain values reflect regional systolic function, and global strain values estimate overall LV systolic function.10 Peak systolic SR correlates well with indices of systolic function and with LV contractility.11,12 Myocardial deformation has been tracked by use of tissue Doppler methods that record Doppler frequency shifts from targets of moving myocardium. It is also possible to measure deformation via 2-D echocardiography with specialized software programs that quantify motion of myocardial reflectors or speckles. Although software algorithms for both Doppler and 2-D methods are different, both methods can provide estimates of regional or segmental myocardial strain and SR.
The assessment of myocardial deformation via 2-D speckle-tracking echocardiography is a relatively new approach for assessment of regional and global LV function. This technique has been used in the evaluation of various cardiac disorders including hypertrophic car-diomyopathy,13 pulmonary hypertension,14 and regional ischemia in humans.15–18 Tissue Doppler ultrasonography—based methods of strain analysis have been used for evaluation of clinically normal dogs19 and dogs with spontaneous dilated cardiomyopathy.20 Strain analysis with 2-D speckle-tracking echocardiography has been studied in healthy unsedated dogs.21
The wider acceptance of myocardial deformation imaging in veterinary patients will eventually require clinical evidence that these indices offer diagnostic information beyond that available via conventional echocardiography. Initial steps in assessing a new echocardiographic method include validation of the index to a gold standard, comparison of the determined variable with variables assessed via conventional methods, assessment of biologic and examiner repeatability of the new echocardiographic method, and a demonstration of the clinical usefulness and uniqueness of the new method. The objective of the study reported here was to compare echocardiographic indices of myocardial strain with invasive measurements of LV systolic function in anesthetized healthy dogs. To this end, invasive measurements of LV contractile function and myocardial strain were simultaneously recorded in clinically normal dogs during various hemodynamic and inotropic conditions. Futhermore, we compared the new myocardial strain indices with conventional 2-D, M-mode, and Doppler echocardiographic variables of LV systolic function. Our hypothesis was that 2-D analysis of longitudinal and radial myocardial strain and SR could provide data that predict LV contractility (ie, LV+dp/dtmax) derived by use of an invasive gold-standard method over a range of hemodynamic states in healthy dogs, and that these new echocardiographic indices would provide additional information about LV function, compared with information provided by use of conventional measures.
Materials and Methods
This study protocol was approved by The Ohio State University Institutional Animal Care and Use Committee. Procedures met the requirements of the US Public Health Service Guide for the Care and Use of Laboratory Animals.
Animals—Seven healthy purpose-bred mature Foxhounds were studied. All dogs underwent a cardiovascular examination involving auscultation and 2-D M-mode Doppler echocardiography without sedation; no abnormalities were detected in any of the 7 dogs.
General anesthesia, cardiac catheterization, and hemodynamic evaluation—Morphinea (0.4 mg/kg, IM) was administered as premedication to each dog. A peripheral 18-gauge over-the-needle catheter was placed in a cephalic vein. Anesthesia was induced with propofolb (4 to 6 mg/kg, IV). Following intubation, each dog was administered isofluranec (1% to 4% isoflurane with 100% oxygen) to maintain a stable plane of anesthesia. Routine anesthetic monitoring included continuous ECG, capnography, and pulse oximetry. Cefazolin sodiumd (22 mg/kg, IV) was administered at the time of induction of anesthesia as well as every 2 hours during the experimental period.
Each dog was placed in left lateral recumbency for surgical instrumentation. The right lateral neck region was clipped, surgically prepared, and draped. An approximately 4- to 5-cm surgical cutdown was performed over the right jugular furrow to expose the right jugular vein and right carotid artery. Infiltration of lidocaine hydrochloridee was used for additional local anesthesia. A 9F or 10F introducer sheath was placed in the jugular vein for right-sided heart instrumentation. A 7.5F Swan-Ganz catheterf was advanced across the sheath into the main pulmonary artery with fluoroscopic guidance. Pressure recordings from the Swan-Ganz catheter were calibrated against pressure recordings obtained by use of a mercury manometer. With this instrumentation, RAPm and systolic, diastolic, and mean pulmonary artery pressures were measured. The right carotid artery was isolated via blunt dissection, and a 9F introducer sheath was placed in the carotid artery. A 100-cm, 8F, high-fidelity, dual-micromanometer catheterg was advanced with fluoroscopic guidance through the vascular sheath within the carotid artery into the aorta and the LV. The distal sensor was positioned at the mid-ventricle level, and the proximal sensor was positioned within the aortic root or ascending aorta. The dual-micromanometer catheter was externally calibrated against a mercury manometer in a saline (0.9% NaCl) solution bath at room temperature (approx 21°C) prior to placement within the body. Once positioned within the LV, the dual-micromanometer catheter was interfaced with a physiologic digital pressure recording systemh and pressure amplifiers. Aortic and LV pressures were recorded, and systolic, diastolic, and mean aortic pressures as well as peak systolic LV pressure and LVEDP were obtained. From the peak systolic LV pressure curve, LV+dp/dtmax was calculated by use of softwarei that was integral to the pressure recording system.
The Swan-Ganz catheter was interfaced to a thermodilution cardiac output meter.j Cardiac output was measured by use of a thermodilution technique as previously described.22–24 Briefly, 3 mL of a cold 5% dextrose solution was injected rapidly into a proximal catheter port located within the right atrium. Resultant changes in blood temperature were measured downstream by use of a thermistor at the tip of the catheter that was located within the main pulmonary artery. Values for cardiac output were discarded when the associated software-generated curve was not of adequate quality. The mean of 3 to 5 measurements was calculated to obtain thermodilution cardiac output and stroke volume values for each hemodynamic state. For each hemodynamic state, cardiac index and SVI were calculated by dividing the mean cardiac output and stroke volume values, respectively, by body surface area.
All measurements were recorded during stable hemodynamic states (< 10% change in measured variables following a minimum interval of 5 minutes of equilibration) and at end expiration. Six hemodynamic states were induced in a repeated-measures, full crossover experiment. Each dog was evaluated under 6 conditions as follows: baseline (consisting of a light plane of anesthesia [1% to 3% isoflurane]), infusion of a low dose of esmolol hydrochloridek (50 μg/kg/min) to create a condition of myocardial depression, infusion of a high dose of esmolol (100 μg/kg/min) to create a more exaggerated condition of decreased inotropy, continued high-dose esmolol infusion with rapid IV infusion of hetastarchl to achieve a mean pulmonary artery pressure of 18 to 20 mm Hg (ie, increased preload), discontinuation of esmolol infusion and low-dose dobutamine hydrochloridem infusion (2.5 μg/kg/min) to restore myocardial function, and infusion of a high dose of dobutamine (5.0 μg/kg/min) to further increase inotropy. All dogs were given furosemiden (2 mg/kg, IV) following the sixth hemodynamic state to decrease LV filling pressures prior to recovery from anesthesia. Infusion of hetastarch was accomplished by use of a rapid infusion fluid pump,o and all drug infusions were administered with a syringe pump.p
Following completion of the experiment, all catheters were removed and the jugular vein and carotid artery were ligated. Standard surgical closure of the incision was performed, and a neck bandage was placed. To prevent development of postoperative pain, buprenorphine HClq (15 μg/kg) was administered SC every 8 hours for 24 hours. Following recovery, 5 dogs were adopted and 2 dogs were transferred to another protocol within the authors' institution.
Conventional and Doppler echocardiography—Standard transthoracic echocardiography25 with continuous ECG was performed during the hemodynamic recording period of each of the 6 hemodynamic states by a trained investigator (NMC). Echocardiographic examination was performed by use of an echocardiographic systemr and a phased-array transducer with a 3.5-MHz nominal frequency. Dogs remained in left lateral recumbency for all imaging acquisitions. Recordings were obtained during periods of sinus rhythm (as determined via ECG) and at end expiration. Raw digitals echocardiographic data were recorded for subsequent analysis on the digital workstation. All measurements were made off-line by the same investigator (NMC), and the mean of individual measurements from 3 to 5 closely timed cardiac cycles was calculated. The investigator making the measurments was unaware of which dog was being evaluated or which hemodynamic state was established. Measurement of the preceding R-R interval on the ECG tracing provided heart rate calculations.
Standard linear, area, and volumetric echocardiographic indices of LV systolic function (ie, fractional shortening, shortening area, and ejection fraction, respectively) were obtained. Images of the LV were obtained at the level of the papillary muscles from the right parasternal short-axis view. Because each dog remained in left lateral position for the duration of the experimental period, this view was obtained from the nondependent side of the thorax. M-mode measurements of the LV diastolic internal diameter (LVID) and LV systolic internal diameter (LVID) were recorded. From these data, LV fractional shortening (FS) was calculated from the following formula: FS = ([LVIDd - LVIDs]/LVIDd) × 100. In the same view of the LV, measurements of the LV diastolic area (LVAd) and LV systolic area (LVAs) were obtained. From these values, LV shortening area (SA) was calculated as follows: SA = ([LVAd − LVAs]/LVAd) × 100. From the right parasternal long-axis view, LV ejection fracton was obtained from images that were optimized for the LV inflow tract. Ejection fraction (EF) was calculated by use of the single-plane Simpson rule disc summation method. End-diastolic volume (LVvold) and end-systolic volume (LVvols) of the LV were calculated, and EF was determined by use of a formula as follows: EF = ([LVvold-LVvols]/LVvold) × 100.
Continuous wave Doppler ultrasonography was used to obtain an aortic velocity flow signal from the left parasternal 3- or 5-chamber view. The sweep speed was optimized to 200 mm/s for all measurements. The LV PEP was measured from the onset of the Q wave on the ECG trace to the onset of the aortic flow signal. The LV ET was measured from the beginning of the aortic flow signal to the onset of the aortic valve closure signal. From these measurements, the PEP:ET ratio was calculated.
2-D echocardiographic strain analysis—As for the conventional echocardiographic and Doppler ultrasonographic imaging, all recordings were acquired during the hemodynamic recording period of each of the 6 hemodynamic states by the same investigator (NMC), who used the same echocardiographic system and transducer as described for the conventional echocardiography.
Good-quality 2-D images of the LV were obtained from the right parasternal short-axis view at the level of the papillary muscles and from the left parasternal apical 4-chamber view. All images were stored digitally in raw digitals format for later off-line analysis. Recordings were made with a frame rate of 60 to 90 frames/s and with 2-D gain and image quality optimized to avoid ultrasonographic artifact and areas of echocardiographic dropout. Event timing for aortic valve closure was recorded from continuous wave Doppler recordings of aortic outflow.
Strain and SR measurements were made off-line on a workstation with 2-D strain software.t One cardiac cycle was defined as starting at the peak of the R wave and ending at the peak of the following R wave. For each view of the LV, the region of interest was defined by tracing the endocardial border of the myocardium. The region of interest was then adjusted to incorporate the entire myocardial thickness. Individual LV segments were considered acceptable for analysis if the tracking quality was deemed adequate by the software and on visual inspection. Strain and SR were calculated for each myocardial segment by use of the software's algorithm for tracking multiple points over a cardiac cycle. Reflected ultrasound beams create an acoustic backscatter, which results in a unique myocardial speckle pattern. Because this pattern remains relatively stable throughout the cardiac cycle, it can be followed frame by frame with the computer software. During the cardiac cycle, displacement of the speckles relative to each other within the pattern represents myocardial deformation. The software algorithm then calculates the displacement, myocardial velocity, strain, and SR for each defined myocardial segment.10,26 The values for these variables are mean values for each given segment. Longitudinal global strain is calculated from the entire region of interest (considered as a single segment), which does not necessarily equal the mean strain for the individual segments.10
One investigator (NMC) measured 6 myocardial segments for each view of the LV that were analyzed by use of 2-D echocardiographic strain analysis. Longitudinal segments were divided into basal septal, midseptal, apical septal, apical lateral, midlateral, and basal lateral segments (Figure 1). Radial segments were designated as anterior septal, anterior, lateral, posterior, inferior, and septal segments (Figure 2). Following calculation of strain and SR curves for each segment, each curve was visually inspected. Software designation of peak strain and SR values from generated curves were verified and corrected as necessary. Peak strain and SR were defined as the maximal systolic value on the respective curve prior to aortic valve closure. Postsystolic peaks—those occurring after aortic valve closure—were not measured. Global strain values were obtained by placement of the cursor on the peak of the global strain curve prior to aortic valve closure. All strain measurements were made by the investigator who was unaware of the results of the hemodynamic recordings.
Representative 2-D echocardiographic image obtained from an anesthetized healthy dog to illustrate a left apical 4-chamber view in which longitudinal segments have been designated as basal septal (yellow), midseptal (light blue), apical septal (green), apical lateral (purple), midlateral (dark blue), and basal lateral (red) segments for strain analysis. LA = Left atrium.
Citation: American Journal of Veterinary Research 72, 5; 10.2460/ajvr.72.5.650
Representative 2-D echocardiographic image obtained from an anesthetized healthy dog to illustrate a right parasternal short-axis view of the LV at the level of the papillary muscles in which radial segments have been designated as anterior septal (yellow), anterior (light blue), lateral (green), posterior (purple), inferior (dark blue), and septal (red) segments for strain analysis. RV = Right ventricle.
Citation: American Journal of Veterinary Research 72, 5; 10.2460/ajvr.72.5.650
Statistical analysis—Individual data points represented the mean of 3 to 5 measurements from closely timed cardiac cycles. All data sets were tested for normality by use of the D'Agostino-Pearson omnibus test. General linear relationships between predictive echocardiographic variables and LV+dp/dtmax were evaluated by inspection of scatterplots prior to linear regression analysis. Descriptive statistics (mean, median, SEM, and SD) were calculated for all echocardiographic and hemodynamic variables.
The Spearman correlation coefficient between heart rate and global strain was determined. The main goal of the study was to investigate relationships (correlations) between global and regional echocardiographic measures of LV systolic function and invasively measured LV+dp/dtmax. Accordingly, statistical analyses were not focused on hemodynamic differences among the induced hemodynamic states, but were focused on prediction of LV+dp/dtmax by use of conventional echocardiographic and 2-D echocardiographic strain indices. These relationships were assessed by use of repeated-measures linear regression, a form of multiple linear regression, as described by Glantz and Slinker.27
A general linear model was used to predict invasively measured LV function from the echocardiographic variables. To account for repeated measurements in dogs, each dog was treated as an independent variable and coded with dummy variables. The model was as follows:
where Bo is a constant (y-intercept), B is a coefficient, IV is the independent (echocardiographic) variable of interest, and D1 through D6 are dummy variables coding for the 7 dogs.27 The dummy variables were included to account for the influence of repeated measurements made in the dogs. The zero-order correlation coefficient was calculated from the multiple linear regression and related the echocardiographic variable of interest to LV+dp/dtmax without adjustment for other independent variables (ie, dogs). The predictive value of each echocardiographic variable was further isolated by calculating partial and part (semipartial) correlation coefficients. The partial correlation coefficient indicates the correlation that remains between a dependent and independent variable after removing the correlation attributable to their mutual association with the other independent variables. The semipartial correlation coefficient relates the dependent variable (LV+dp/dtmax) to the predictive (echocardiographic) variable after the linear effects of the other independent variables in the model have been removed.28 Variance inflation factors were calculated to identify colinearity between independent variables. For all analyses, a value of P < 0.05 was considered significant. Statistical analyses were performed by use of commercial statistical software packages.u-w
Results
Animals—Dogs were 1 to 5 years old (mean age, 2.2 years) and weighed 21 to 26.6 kg (mean weight, 24.0 kg). For study purposes, dogs were designated as dogs 1 through 7. Four dogs were sexually intact males, and 3 dogs were sexually intact females.
Neither echocardiographic nor hemodynamic data were collected for dog 2 during low-dose dobutamine treatment because there was no discernible change in heart rate, cardiac output, or intravascular pressures at the lower dose of dobutamine relative to the data obtained during the preceding hemodynamic state. Hemodynamics were altered by the high-dose dobutamine treatment, and those data were included in analysis.
Cardiac catheterization—Data for invasive cardiac hemodynamics from the 7 dogs during the 6 hemodynamic states were summarized (Table 1). Alterations in LV contractile function were tracked as changes in LV+dp/dtmax. A decrease in LV+dp/dtmax was evident following low-dose esmolol treatment. High-dose esmolol treatment further depressed LV function, as supported by an additional decrease in LV+dp/dtmax. The addition of hetastarch to the high-dose esmolol treatment had a minimal effect on LV+dp/dtmax. Low-dose dobutamine treatment returned LV+dp/dtmax to the baseline value, whereas high-dose dobutamine treatment increased LV+dp/dtmax further, approximately doubling the value observed with low-dose esmolol treatment.
Hemodynamic variables (mean ± SD) in 7 anesthetized healthy dogs before (baseline) and during 5 treatments* administered in sequence to induce 6 hemodynamic states.
| Variable | Baseline | Low-dose esmolol | High-dose esmolol | High-dose esmolol and hestarch | Low-dose dobutamine | High-dose dobutamine |
|---|---|---|---|---|---|---|
| LV+dp/dtmax (mm Hg/s) | 2,024 ± 434 | 1,561 ± 229 | 1,355 ± 251 | 1,298 ± 186 | 2,084 ± 622 | 3,192 ± 914 |
| Heart rate (beats/min) | 114 ± 18 | 109 ± 14 | 109 ± 12 | 113 ± 13 | 133 ± 15 | 155 ± 15 |
| CI (L/min/m2) | 7.37 ± 2.30 | 6.17 ± 1.80 | 5.90 ± 1.70 | 6.43 ± 1.30 | 11.77 ± 3.26 | 15.91 ± 3.51 |
| SVI (mL/m2) | 64.4 ± 17.6 | 56.5 ± 14.5 | 54.5 ± 14.7 | 57.7 ± 12.7 | 88.0 ± 19.4 | 102.6 ± 18.5 |
| RAPm(mm Hg) | 1.36 ± 5.42 | 4.05 ± 3.70 | 6.47 ± 4.66 | 15.61 ± 3.49 | 12.5 ± 4.6 | 10.59 ± 4.92 |
| LVEDP (mm Hg) | 15.4 ± 4.7 | 18.6 ± 4.9 | 20.6 ± 6.2 | 30.2 ± 5.6 | 33.4 ± 3.6 | 31.5 ± 3.6 |
| Peak LVP (mm Hg) | 116.1 ± 21.0 | 107.1 ± 15.3 | 103.6 ± 14.5 | 103.5 ± 11.8 | 114.7 ± 13.7 | 125.1 ± 17.4 |
| AoPs (mm Hg) | 114.1 ± 23.6 | 105.7 ± 16.9 | 101.3 ± 15.3 | 100.9 ± 11.7 | 107.2 ± 10.5 | 110.2 ± 17.6 |
| AoPd (mm Hg) | 82.8 ± 17.9 | 74.0 ± 15.2 | 71.8 ± 14.7 | 72.5 ± 10.5 | 69.9 ± 7.5 | 62.2 ± 14.4 |
| AoPm (mm Hg) | 97.6 ± 21.4 | 88.8 ± 17.1 | 85.3 ± 15.9 | 86.0 ± 11.7 | 88.7 ± 9.7 | 84.3 ± 15.5 |
| PAPs (mm Hg) | 21.9 ± 4.2 | 21.6 ± 5.2 | 23.2 ± 5.5 | 31.4 ± 6.1 | 35.5 ± 5.8 | 43.8 ± 6.7 |
| PAPd (mm Hg) | 12.5 ± 4.8 | 11.7 ± 3.1 | 13.0 ± 2.0 | 21.5 ± 5.6a | 20.3 ± 3.7 | 25.7 ± 7.7 |
| PAPm (mm Hg) | 16.8 ± 4.2 | 16.3 ± 3.8 | 17.4 ± 3.3 | 26.1 ± 5.8 | 27.2 ± 3.4 | 33.5 ± 7.2 |
Each dog was evaluated under 6 conditions as follows: baseline (consisting of a light plane of anesthesia [1 % to 3% isoflurane]), infusion of a low dose of esmolol hydrochloride (50 μg/kg/min) to create a condition of myocardial depression, infusion of a high dose of esmolol (100 μg/kg/min) to create a more exaggerated condition of decreased inotropy, continued high-dose esmolol infusion with rapid IV infusion of hetastarch to achieve a mean pulmonary artery pressure of 18 to 20 mm Hg (ie, increased preload), discontinuation of esmolol infusion and low-dose dobutamine hydrochloride infusion (2.5 μg/kg/min) to restore myocardial function, and infusion of a high dose of dobutamine (5.0 μg/kg/min) to further increase inotropy. All measurements were recorded during stable hemodynamics (, 10% change in measured variables following a minimum interval of 5 minutes of equilibration) and at end expiration.
AoPd = Diastolic aortic pressure. AoPm = Mean aortic pressure. AoPs = Systolic aortic pressure. CI = Cardiac index. LVP = Peak LV pressure. PAPd = Diastolic pulmonary arterial pressure. PAPm = Mean pulmonary arterial pressure. PAPs = Systolic pulmonary arterial pressure.
Left- and right-sided heart filling pressures were estimated by the combination of LVEDP, diastolic pulmonary artery pressure, and RAPm. Filling pressures were increased mildly with both esmolol treatments, as evidenced by an increase from baseline in LVEDP and RAPm. The addition of hetastarch to the high-dose esmolol treatment markedly increased all markers of filling pressures; LVEDP increased to approximately double the baseline value. High- and low-dose dobutamine treatments had little effect on filling pressures in hetastarch-loaded dogs.
Heart rate and SVI decreased with low-dose esmolol treatment but remained unchanged with high-dose esmolol treatment. The addition of hetastarch to the high-dose esmolol treatment mildly increased both heart rate and SVI. Further increases in these indices were seen with low-dose dobutamine treatment, but high-dose dobutamine treatment resulted in marked increases in these variables.
Conventional echocardiography—Data for conventional echocardiographic indices of LV systolic function for each of the 6 hemodynamic states were obtained (Table 2). As expected, conventional indices of LV systolic function followed a similar pattern of response to the various treatments as that of the changes in LV+dp/dtmax. All indices indicated depression of LV systolic function with low-dose or high-dose esmolol treatments. The addition of hetastarch to the high-dose esmolol treatment mildly improved the PEP:ET ratio but had minimal effect on the other indices. All indices indicated improvement in LV systolic function with low-dose dobutamine treatment and further increments in systolic function following high-dose dobutamine treatment.
Calculated (mean ± SD) conventional echocardiographic variables of LV systolic function in 7 anesthetized healthy dogs before (baseline) and during 5 treatments* administered in sequence to induce 6 hemodynamic states.
| Variable | Baseline | Low-dose esmolol | High-dose esmolol | High-dose esmolol and hestarch | Low-dose dobutamine | High-dose dobutamine |
|---|---|---|---|---|---|---|
| FS (%) | 33.8 ± 4.0 | 28.9 ± 6.5 | 26.3 ± 10.2 | 23.5 ± 4.9 | 32.8 ± 7.0 | 43.0 ± 8.3 |
| SA (%) | 46.9 ± 5.5 | 42.8 ± 9.5 | 38.8 ± 13.4 | 40.5 ± 5.8 | 51.0 ± 10.5 | 63.7 ± 9.6 |
| EF (%) | 54.7 ± 6.7 | 49.8 ± 7.8 | 45.1 ± 11.0 | 47.7 ± 7.4 | 61.5 ± 10.2 | 74.6 ± 6.4 |
| PEP:ET ratio | 0.31 ± 0.063 | 0.34 ± 0.068 | 0.36 ± 0.071 | 0.30 ± 0.051 | 0.23 ± 0.022 | 0.23 ± 0.083 |
EF = Ejection fraction. FS = Fractional shortening. SA = Shortening area.
See Table 1 for remainder of key.
2-D echocardiographic strain analysis—All of the 504 myocardial segments examined via 2-D speckle-tracking echocardiography were used for quantitative data analysis. As expected, values for longitudinal peak systolic strain and peak systolic SR were negative and values for radial peak systolic strain and SR were positive for all dogs in each hemodynamic state (Figure 3; Tables 3 and 4). There was considerable heterogeneity of strain and SR values across the various myocardial segments, especially among the longitudinal values. In general, esmolol infusion reduced the deformation indices, the addition of hetastarch had minimal effects, and the infusion of dobutamine markedly enhanced strain and SR. The Spearman correlation coefficient between heart rate (during longitudinal strain measurement) and global longitudinal strain was −0.293 (P = 0.66).
Longitudinal strain and SR values for regional LV segments determined via 2-D speckle-tracking echocardiography in 7 anesthetized healthy dogs before (baseline) and during 5 treatments* administered in sequence to induce 6 hemodynamic states.
| Variable | Segment† | Baseline | Low-dose esmolol | High-dose esmolol | High-dose esmolol and hestarch | Low-dose dobutamine | High-dose dobutamine |
|---|---|---|---|---|---|---|---|
| Longitudinal strain (%) | Basal septal | −12.64 ± 2.62 | −11.29 ± 2.39 | −9.30 ± 2.67 | −10.95 ± 1.77 | −14.34 ± 2.15 | −18.08 ± 1.69 |
| Midseptal | −16.60 ± 3.00 | −14.48 ± 3.53 | −13.17 ± 4.17 | −14.22 ± 1.49 | −20.34 ± 2.21 | −23.95 ± 2.58 | |
| Apical septal | −23.20 ± 4.38 | −17.84 ± 5.07 | −18.04 ± 5.80 | −18.46 ± 2.63 | −27.90 ± 4.31 | −28.87 ± 2.46 | |
| Apical lateral | −22.97 ± 5.18a | −17.33 ± 3.96 | −16.75 ± 5.53 | −14.20 ± 3.92 | −24.35 ± 4.69 | −23.58 ± 3.18b | |
| Midlateral | −16.18 ± 2.01c | −14.26 ± 2.02 | −12.83 ± 3.07 | −13.45 ± 2.67 | −17.70 ± 4.20 | −19.97 ± 3.48 | |
| Basal lateral | −10.91 ± 1.36 | −10.56 ± 1.36 | −9.59 ± 2.51 | −11.76 ± 2.64 | −14.88 ± 2.76 | −20.79 ± 4.07d | |
| Longitudinal SR(1/s) | Basal septal | −1.01 ± 0.20 | −0.78 ± 0.20 | −0.70 ± 0.17 | −0.73 ± 0.14 | −1.04 ± 0.09 | −1.72 ± 0.58 |
| Midseptal | −1.52 ± 0.27 | −1.17 ± 0.26 | −1.08 ± 0.32 | −0.99 ± 0.11 | −1.99 ± 0.20 | −2.67 ± 0.44 | |
| Apical septal | −2.29 ± 0.41 | −1.49 ± 0.40 | −1.52 ± 0.52 | −1.30 ± 0.18 | −2.89 ± 0.51e | −3.43 ± 0.54 | |
| Apical lateral | −2.03 ± 0.38 | −1.35 ± 0.35 | −1.42 ± 0.46 | −1.14 ± 0.28 | −2.43 ± 0.54 | −2.72 ± 0.67 | |
| Midlateral | −1.31 ± 0.25 | −0.97 ± 0.24 | −0.87 ± 0.26 | −0.90 ± 0.20 | −1.54 ± 0.41f | −1.99 ± 0.45 | |
| Basal lateral | −1.02 ± 0.18 | −0.78 ± 0.13 | −0.74 ± 0.12 | −0.90 ± 0.13 | −1.42 ± 0.18 | −2.22 ± 0.75 | |
| Global strain (%) | −16.61 ± 2.21 | −13.84 ± 2.61 | −12.99 ± 3.66 | −14.02 ± 1.52 | −19.94 ± 2.81 | −22.04 ± 2.18 | |
| Heart rate (beats/min) | 116 ± 19 | 109 ± 12 | 108 ± 12 | 114 ± 13 | 138 ± 15 | 158 ± 12 |
Six longitudinal myocardial segments were analyzed via 2-D echocardiographic strain analysis. Longitudinal segments were designated as basal septal, midseptal, apical septal, apical lateral, midlateral, and basal lateral segments.
During this treatment period, apical lateral segment data failed normality testing (median, −24.620; 25th and 75th percentiles, −25.660 and −18.138, respectively).
During this treatment period, apical lateral segment data failed normality testing (median, −22.667; 25th and 75th percentiles, −23.590 and −22.343, respectively).
During this treatment period, midlateral segment data failed normality testing (median, −16.477; 25th and 75th percentiles, −17.437 and −16.163, respectively).
During this treatment period, apical lateral segment data failed normality testing (median, −21.193; 25th and 75th percentiles, −23.883 and −20.867, respectively).
During this treatment period, apical septal segment data failed normality testing (median, −3.047; 25th and 75th percentiles, −3.083 and −2.947, respectively).
During this treatment period, midlateral segment data failed normality testing (median, −1.523; 25th and 75th percentiles, −1.537 and −1.367, respectively).
See Table 1 for remainder of key.
Radial strain and SR values for regional LV segments determined via 2-D speckle-tracking echocardiography in 7 anesthetized healthy dogs before (baseline) and during 5 treatments* administered in sequence to induce 6 hemodynamic states.
| Variable | Segment† | Baseline | Low-dose esmolol | High-dose esmolol | High-dose esmolol | Low-dose dobutamine and hetastarch | High-dose dobutamine |
|---|---|---|---|---|---|---|---|
| Radial strain (%) | Anterior septal | 36.76 ± 7.42a | 31.75 ± 15.16 | 32.15 ± 15.04 | 28.87 ± 3.35 | 41.10 ± 9.28 | 54.16 ± 8.10e |
| Anterior | 41.87 ± 6.09 | 35.30 ± 11.80 | 38.21 ± 16.08 | 35.22 ± 4.58 | 52.80 ± 9.83 | 58.36 ± 5.82 | |
| Lateral | 48.64 ± 5.64 | 39.69 ± 10.67 | 41.27 ± 16.34 | 43.37 ± 8.91 | 63.43 ± 12.39 | 64.22 ± 7.34 | |
| Posterior | 52.62 ± 7.08 | 41.22 ± 10.02 | 40.36 ± 16.03 | 46.38 ± 12.43 | 65.30 ± 15.30 | 67.82 ± 9.49 | |
| Inferior | 50.69 ± 8.69 | 38.96 ± 12.15 | 36.17 ± 14.56 | 43.81 ± 14.80 | 58.83 ± 16.60 | 66.92 ± 11.53 | |
| Septal | 43.23 ± 7.70 | 34.49 ± 16.05 | 31.55 ± 13.80 | 35.49 ± 11.24 | 46.34 ± 13.87 | 61.39 ± 14.14 | |
| Radial SR (1/s) | Anterior septal | 1.88 ± 0.40 | 1.50 ± 0.48 | 1.66 ± 0.62 | 1.59 ± 0.20 | 2.28 ± 0.56 | 3.50 ± 0.85 |
| Anterior | 2.25 ± 0.40 | 1.74 ± 0.47c | 2.10 ± 1.01 | 1.80 ± 0.35 | 2.70 ± 0.88d | 4.12 ± 1.03 | |
| Lateral | 2.38 ± 0.42 | 1.88 ± 0.50 | 2.22 ± 1.05 | 2.00 ± 0.38 | 2.92 ± 0.92 | 4.51 ± 1.03 | |
| Posterior | 2.17 ± 0.31 | 1.72 ± 0.42 | 1.92 ± 0.74 | 1.99 ± 0.35 | 2.93 ± 0.89 | 4.27 ± 0.86 | |
| Inferior | 2.15 ± 0.35 | 1.61 ± 0.38 | 1.61 ± 0.43 | 1.89 ± 0.35 | 2.75 ± 0.80 | 3.77 ± 0.76 | |
| Septal | 2.24 ± 0.78b | 1.52 ± 0.50 | 1.57 ± 0.44 | 1.73 ± 0.33 | 2.42 ± 0.41 | 3.34 ± 0.89 | |
| Heart rate (beats/min) | 114 ± 18 | 109 ± 14 | 109 ± 12 | 113 ± 13 | 133 ± 15 | 155 ± 15 |
Six radial myocardial segments were analyzed via 2-D echocardiographic strain analysis. Radial segments were designated as anterior septal, anterior, lateral, posterior, inferior, and septal segments.
During this treatment period, anterior septal segment data failed normality testing (median, 36.343; 25th and 75th percentiles, 31.657 and 37.162, respectively).
During this treatment period, anterior septal segment data failed normality testing (median, 58.472; 25th and 75th percentiles, 43.767 and 59.833, respectively).
During this treatment period, anterior segment data failed normality testing (median, 1.673; 25th and 75th percentiles, 1.472 and 1.730, respectively).
During this treatment period, anterior segment data failed normality testing (median, 2.323; 25th and 75th percentiles, 2.117 and 3.743, respectively).
During this treatment period, septal segment data failed normality testing (median, 1.927; 25th and 75th percentiles, 1.838 and 2.366, respectively).
See Table 1 for remainder of key.
Representative curves of LV longitudinal strain (A), longitudinal SR (B), radial strain (C), and radial SR (D) obtained via 2-D echocardiography from an anesthetized healthy dog during treatment with low-dose dobutamine hydrochloride infusion (2.5 μg/kg/min). For each of the 4 panels, there are multiple tracings, each of which represents a segment of the LV myocardium. The color of each tracing corresponds to the segment identified with the same color in Figures 1 and 2. Values for peak strain and SR were identified as the maximal systolic value on the respective curve. Numbers on the y-axis of panels A and C are values for strain measured in percentage, and numbers on the x-axis of panels B and C are values for SR measured in units of 1 per second; numbers on the x-axis of each panel are milliseconds. AVC = Aortic valve closure.
Citation: American Journal of Veterinary Research 72, 5; 10.2460/ajvr.72.5.650
Repeated-measures linear regression analysis was used to determine correlation coefficients between echocardiographic indices of LV systolic function and invasive LV+dp/dtmax (Table 5). Significant (P < 0.001) correlations were found between LV+dp/dtmax and all conventional echocardiographic indices of LV systolic function and between LV+dp/dtmax and all strain and SR variables. After accounting for the influence of repeated measurements (a between-subjects effect), semipartial regression coefficients were calculated and advanced as the most meaningful estimates for prediction of LV+dp/dtmax from the echocardiographic variables. These semipartial correlation coefficients were −0.773 for global longitudinal strain, ranged from −0.562 to −0.786 for segmental longitudinal strain, and ranged from −0.777 to −0.875 for segmental longitudinal SR (Figure 4). For radial segments, semipartial correlation coefficients ranged from 0.654 to 0.811 for strain and from 0.748 to 0.775 for SR. Correlations of traditional echocardiographic indices with LV+dp/dtmax were also significant; the lowest correlation coefficient was that between LV+dp/dtmax and the PEP:ET ratio (−0.586), and the highest value correlation coefficient was that between LV+dp/dtmax and the 2-D short-axis shortening area (0.821). Semipartial correlation coefficients between global longitudinal strain and traditional echocardiographic indices of LV systolic function were also significant (P < 0.001) and ranged from −0.656 for the shortening fraction to −0.726 and −0.744 for the shortening area and ejection fraction, respectively.
Correlation coefficients* for prediction of LV+dp/dtmax from 2-D speckle-tracking echocardiographic variables determined in 7 anesthetized healthy dogs.
| Echocardiographic variable | Zero-order correlation | Partial correlation | Semipartial correlation |
|---|---|---|---|
| LV shortening fraction | 0.783 | 0.880 | 0.812 |
| LV fractional area shortening | 0.760 | 0.890 | 0.821 |
| LV ejection fraction | 0.774 | 0.870 | 0.802 |
| ET during PEP | −0.608 | −0.636 | −0.586 |
| LV longitudinal SR by segment | |||
| Basal septal | −0.818 | −0.873 | −0.805 |
| Midseptal | −0.814 | −0.843 | −0.777 |
| Apical septal | −0.846 | −0.872 | −0.804 |
| Apical lateral | −0.833 | −0.869 | −0.802 |
| Midlateral | −0.898 | −0.949 | −0.875 |
| Basal lateral | −0.870 | −0.911 | −0.840 |
| LV longitudinal global strain | −0.811 | −0.838a | −0.773b |
| LV longitudinal strain by segment | |||
| Basal septal | −0.714 | −0.827 | −0.763 |
| Midseptal | −0.755 | −0.827 | −0.763 |
| Apical septal | −0.702 | −0.723 | −0.667 |
| Apical lateral | −0.629 | −0.609 | −0.562 |
| Midlateral | −0.793 | −0.841 | −0.776 |
| Basal lateral | −0.720 | −0.852 | −0.786 |
| LV radial strain by segment | |||
| Anterior septal | 0.685 | 0.825 | 0.761 |
| Anterior | 0.640 | 0.721 | 0.665 |
| Lateral | 0.623 | 0.709 | 0.654 |
| Posterior | 0.643 | 0.742 | 0.684 |
| Inferior | 0.667 | 0.792 | 0.731 |
| Septal | 0.709 | 0.879 | 0.811 |
| LV radial SR by segment | |||
| Anterior septal | 0.804 | 0.817 | 0.754 |
| Anterior | 0.767 | 0.816 | 0.752 |
| Lateral | 0.783 | 0.832 | 0.769 |
| Posterior | 0.792 | 0.840 | 0.775 |
| Inferior | 0.791 | 0.833 | 0.768 |
| Septal | 0.788 | 0.811 | 0.748 |
The zero-order correlation coefficient is the coefficient from the repeated linear regression for the echocardiographic predictor (unadjusted for other independent variables). The partial correlation coefficient shows the correlation that remains between a dependent and independent variable after removing the correlation due to their mutual association with the other independent variables. The semipartial correlation coefficient relates the dependent variable (LV+dp/dtmax) to the predictive (echocardiographic) variable after the linear effects of the other independent variables in the model have been removed.
All correlations were significant (P < 0.001).
Coefficients for global strain with removal of 1 outlier with standardized residual of 3.736.
Scatterplots of LV+dp/dtmax (mm Hg/s) versus global longitudinal strain (A) or segmental SR (B) determined in 7 anesthetized healthy dogs in a study to compare 2-D echocardiographic indices of myocardial strain with invasive measurements of LV systolic function. The data points represent the mean values obtained for each of the plotted indices for each dog during the 6 hemodynamic states. The outlined data point in panel A is a statistical outlier. In panel B, the SR was obtained from the middle lateral longitudinal segment, the segment that had the highest semipartial correlation with LV+dp/dtmax.
Citation: American Journal of Veterinary Research 72, 5; 10.2460/ajvr.72.5.650
Discussion
Assessment of LV systolic function in patients with heart disease is routinely performed via echocardiography in the clinical setting. One major limitation of conventional echocardiographic indices of systolic function relates to their dependence on both preload and afterload.5,6 Invasive indices of LV contractile function have traditionally been considered the gold standard for assessment of LV systolic function. Although LV+dp/dtmax has less dependence on afterload and LV elastance is thought to be nearly load independent, assessments of these variables are neither practical nor economically feasible in patients. Tagged magnetic resonance imaging is considered by many to be the absolute noninvasive gold standard for measurement of LV contractile function. However, this expensive method requires anesthesia, which depresses contractility in higher-risk canine patients.
Several studies15–18,29–34 in people and other animals have revealed the potential of 2-D echocardiographic strain imaging for noninvasive assessment of LV systolic function in both healthy individuals and patients with cardiac disease. Although tissue Doppler imaging-based strain analysis has been performed in clinically normal dogs19 and dogs with dilated cardiomyopathy,20 2-D echocardiographic-based techniques require further study and validation for assessment of cardiac function in clinical canine patients.
The results of the present study support the feasibility of 2-D speckle-tracking echocardiographic-based strain analysis of LV systolic function over a wide range of inotropic and hemodynamic states in anesthetized healthy dogs. It was possible to evaluate all LV segments with 2-D echocardiographic strain software. In human studies, the success rate for 2-D echocardiographic-based strain analysis of segments was high; in clinically normal individuals, 98%31 and 97.8%35 of segments were analyzed. Among the 6 analyzed segments (or regions) measured in the present study, variability for longitudinal strain indices was greater than it was for radial strain indices, but longitudinal strain indices were better correlated overall with LV+dp/dtmax.
As expected, significant linear correlations were identified between LV+dp/dtmax and conventional echocardiographic indices of global LV systolic function in the present study. A similar correlation was identified with strain imaging, and longitudinal SR had the strongest correlations with LV+dp/dtmax. This finding is consistent with results of previous studies,11,12,36,37 which indicated that reported peak systolic SR correlates well to load-independent indices of systolic function and therefore to LV contractility. In the present study, longitudinal global strain also showed a strong correlation with LV+dp/dtmax (r = −0.773). This variable has some promise as a single rapid, noninvasive index of LV systolic function. Correlations between radial strain or SR and LV+dp/dtmax were weaker than the correlations between longitudinal strain or SR and LV+dp/dtmax, but the radial plane measurements were much more consistent across segments. Although a linear model could be fitted to the data by use of the repeated-measures regression model, it was also possible to demonstrate a significant quadratic relationship between variables. More studies are needed to assess the best-fit relationship between LV+dp/dtmax and global longitudinal strain in awake dogs and dogs with overt cardiac disease.
Other investigations in dogs with experimentally induced conditions have revealed strong correlations between results of 2-D echocardiographic strain analysis and various established indices of LV systolic function. In dogs in which changes in preload followed by acute regional ischemia were induced, there was good correlation between tagged magnetic resonance imaging and 2-D echocardiographic strain analysis.31 Similar to findings of the present study, the previous investigation31 identified a stronger correlation (r = 0.90) between long-axis measurements obtained via 2-D echocardiographic strain analysis and those obtained via magnetic resonance imaging, compared with the correlation (r = 0.79) for short-axis variables derived via those 2 methods. In open-thorax pigs with experimentally induced acute ischemia, there is a strong correlation (r = 0.94) between strain values obtained via a 2-D echocardiographic method and those obtained by use of sonomicrometry crystals.29 Exceptionally high correlation (r = 0.99) was identified between results of 2-D echocardiographic strain analysis and strain measurements obtained by use of sonomicrometry crystals in in vitro studies29,30 involving a tissue-mimicking gelatin block. Such high correlation is unlikely to be reproduced in in vivo experiments or in clinical patients.
In the present study, the purpose of the treatments was to induce a range of inotropic and hemodynamic states during which correlations between conventional echocardiographic, 2-D speckle-tracking echocardiographic strain—based, and invasive measures of LV systolic function could be evaluated. Although the baseline period provided a starting point without any treatment effects, inhalant anesthestics cause myocardial depression. The 2 esmolol treatments further depressed LV systolic function, as evidenced by decreases in LV+dp/dtmax and SVI. The combination of hetastarch and high-dose esmolol treatment was included in the protocol to mimic heart failure, a state in which LV systolic function is decreased and elevated LV filling pressures are present. Increasing preload with hetastarch loading had only subtle positive effects in terms of reversing depression of cardiac function. These results suggested that following high-dose esmolol treatment, the LV myocardium was nearly maximally depressed. Although the increase in preload (Frank-Starling mechanism) resulted in marked increases in ventricular filling pressures (eg, LVEDP, diastolic pulmonary artery pressure, and RAPm), this had little effect on LV+dp/dtmax or strain indices. Dobutamine infusion improved LV systolic function and cardiac output (values of LV+dp/dtmax and SVI, respectively) but, as expected, caused only modest reductions in ventricular filling pressures.38
In the present study, both the conventional echocardiographic and 2-D echocardiographic strain-based indices of LV systolic function derived during the changing hemodynamic states tracked global changes well. The alterations in LV systolic function induced in the dogs of the present study were relatively dramatic, and it is impossible to know from these data whether deformation indices would be more sensitive to subtle depression in LV contractility, as might be observed in preclinical dilated cardiomyopathy or advanced mitral valvular regurgitation (2 clinical conditions in which applications of these indices might be potentially helpful).
The present study had several limitations. The small number of dogs limits extrapolation of these findings to larger populations. The investigation included healthy, physically large research dogs and did not include dogs of various sizes or ages or dogs with naturally occuring cardiac disease. Research dogs were chosen because it was not possible to use client-owned dogs for an invasive protocol such as this. The study population was assessed under general anesthesia, and the baseline values obtained for strain and SR should not be used as reference values for those obtained in awake, healthy dogs (for which there are some published data19,21). Another limitation is that during the study protocol, it was not possible to blind the echocardiographer who obtained images to the treatment being administered; however, blinding of the investigator with regard to the specific dog and hemodynamic state during the measurement of the conventional echocardiographic indices presumably compensated for this limitation. The investigator was not unaware of the hemodynamic states of the dogs during strain analysis, but the automated, software-based nature of this technique makes blinding during strain measurements unnecessary. The treatments used in the present study were designed to induce substantial changes in LV systolic function, and the ability of these methods to detect more subtle alterations in LV systolic function will probably require longitudinal studies of dogs at high risk for heart failure and populations of healthy controls evaluated over the same period. Clinical patients may have regional cardiac abnormalities, which could be identified by the segmental approach of strain imaging by use of tissue Doppler or 2-D ultrasonic speckle-tracking methods. The initial assumption that loading conditions would not influence strain measures has been tempered by results of several studies39–41 indicating the opposite effect. Finally, although both LV+dp/dtmax and LV elastance are considered invasive gold standards of global LV function, many researchers prefer the use of elastance measurement to the method used in the present study. Nevertheless, the 2 techniques parallel LV contractility in dogs, and the derived data correlate well.42
The results of the present study indicated that strain and SR derived via 2-D speckle-tracking echocardiography are useful noninvasive indices of LV systolic function over a range of hemodynamic states in anesthetized healthy dogs. Specifically longitudinal SR values had a strong correlation with invasively measured LV+dp/dtmax. Further studies are needed to assess the clinical applicability and measurement variability of this technique in healthy dogs and dogs with naturally occurring cardiac disease.
ABBREVIATIONS
| ET | Ejection time |
| LV | Left ventricle |
| LV+dp/dtmax | Maximum rate of rise (first derivative) of left ventricle systolic pressure |
| LVEDP | End-diastolic left ventricle pressure |
| PEP | Pre-ejection period |
| RAPm | Mean right atrial pressure |
| SR | Strain rate |
| SVI | Stroke volume index |
Morphine, Baxter Healthcare Corp, Deerfield, Ill.
PropoFlo, Abbott Laboratories, North Chicago, Ill.
Isoflurane, Abbott Laboratories, North Chicago, Ill.
Cefazolin sodium for injection USP, Sandoz Inc, Broomfield, Colo.
2% Lidocaine HCl, Abbott Laboratories, North Chicago, Ill.
Swan-Ganz catheter, 7.5F, Baxter Healthcare Corp, Irvine, Calif.
Millar dual pressure transducer catheter, model SPC-780C, Millar Instruments Inc, Houston, Tex.
Model 720-USB, Dataq Instruments, Akron, Ohio.
WINDAQ/PRO and Advanced Codas, Dataq Instruments, Akron, Ohio.
Cardiomax III, Columbus Instruments, Columbus, Ohio.
Esmolol HCl injection, Bedford Laboratories, Bedford, Ohio.
Hetastarch, Hospira Inc, Lake Forrest, Ill.
Dobutamine HCl, Ben Venue Laboratories Inc, Bedford, Ohio.
Furoject, Butler Animal Health Supply, Dublin, Ohio.
Harvard Apparatus Peristaltic Pump, model HA 66, Instech Laboratories Inc, Plymouth Meeting, Pa.
Medfusion 2010i syringe pump, Medex Inc, Duluth, Ga.
Buprenorphine HCl, Bedford Laboratories, Bedford, Ohio.
Vivid 7 Echocardiographic System, General Electric, Waukesha, Wis.
Digital Imaging and Communications in Medicine (DICOM), Rosslyn, Va.
EchoPAC 2D Strain software, q Analysis, version 6.1, General Electric, Waukesha, Wis.
SPSS, version 17.0, SPSS Inc, Chicago, Ill.
SigmaStat, version 3.5, SPSS Inc, Chicago, Ill.
GraphPad Prism, version 5.00 for Windows, GraphPad Software Inc, San Diego, Calif.
References
- 1.
LeWinter MM, Osol G. Normal physiology of the cardiovascular system. In: Fuster V, Alexander RW, O'Rourke RA, eds. Hurst's the heart. 10th ed. New York: McGraw-Hill, 2001; 63–94.
- 2.
Carroll JD, Hess OM. Assessment of normal and abnormal cardiac function. In: Zipes DP, Libby P, Bonow RO, et al, eds. Braunwald's heart disease: a textbook of cardiovascular medicine. 7th ed. Philadelphia: Elsevier Saunders, 2005; 491–507.
- 3.
Chung N, Nishimura RA, Holmes DR Jr, et al. Measurement of left ventricular +dp/dt by simultaneous Doppler echocardiography and cardiac catheterization. J Am Soc Echocardiogr 1992; 5: 147–152.
- 4.
Bargiggia GS, Bertucci C, Recusani F, et al. A new method for estimating left ventricular +dp/dt by continuous wave Doppler-echocardiography. Validation studies at cardiac catheterization. Circulation 1989; 80: 1287–1292.
- 5.
Feigenbaum H, Armstrong WF, Ryan T. Evaluation of systolic and diastolic function of the left ventricle. In: Feigenbaum H, Armstrong WF, Ryan T, eds. Feigenhaum's echocardiography. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2005; 138–180.
- 6.
Otto CM. Left and right ventricular systolic function. In: Otto CM, ed. Textbook of clinical echocardiography. 3rd ed. Philadelphia: Elsevier Saunders, 2004; 131–165.
- 7.
Gilman G, Khandheria BK, Hagen ME, et al. Strain rate and strain: a step-by-step approach to image and data acquisition. J Am Soc Echocardiogr 2004; 17: 1011–1020.
- 8.
Sutherland GR, Di Salvo G, Claus P, et al. Strain and strain rate imaging: a new clinical approach to quantifying regional myocardial function. J Am Soc Echocardiogr 2004; 17: 788–802.
- 9.
Pellerin D, Sharma R, Elliott P, et al. Tissue Doppler, strain, and strain rate echocardiography for the assessment of left and right systolic ventricular function. Heart 2003; 89 (suppl 3): iii9–iii17.
- 10.↑
Teske AJ, De Boeck BW, Melman PG, et al. Echocardiographic quantification of myocardial function using tissue deformation imaging—a guide to image acquisition and analysis using tissue Doppler and speckle tracking. Cardiovasc Ultrasound 2007; 5: 27–45.
- 11.
Weidemann F, Jamal F, Sutherland GR, et al. Myocardial function defined by strain rate and strain during alterations in inotropic states and heart rate. Am J Physiol Heart Circ Physiol 2002; 283: H792–H799.
- 12.
Greenberg NL, Firstenberg MS, Castro PL, et al. Doppler-derived myocardial systolic strain rate is a strong index of left ventricular contractility. Circulation 2002; 105: 99–105.
- 13.↑
Saghir M, Areces M, Makan M. Strain rate imaging differentiates hypertensive cardiac hypertrophy from physiologic cardiac hypertrophy (athlete's heart). J Am Soc Echocardiogr 2007; 20: 151–157.
- 14.↑
Pirat B, McCulloch ML, Zoghbi WA. Evaluation of global and regional right ventricular systolic function in patients with pulmonary hypertension using a novel speckle tracking method. Am J Cardiol 2006; 98: 699–704.
- 15.
Sun JP, Niu J, Chou D, et al. Alterations of regional myocardial function in a swine model of myocardial infarction assessed by echocardiographic 2-dimensional strain imaging. J Am Soc Echocardiogr 2007; 20: 498–504.
- 16.
Winter R, Jussila R, Nowak J, et al. Speckle tracking echocardiography is a sensitive tool for the detection of myocardial ischemia: a pilot study from the catheterization laboratory during percutaneous coronary intervention. J Am Soc Echocardiogr 2007; 20: 974–981.
- 17.
Migrino RQ, Zhu X, Pajewski N, et al. Assessment of segmental myocardial viability using regional 2-dimensional strain echocardiography. J Am Soc Echocardiogr 2007; 20: 342–351.
- 18.
Popovic ZB, Benejam C, Bian J, et al. Speckle-tracking echocardiography correctly identifies segmental left ventricular dysfunction induced by scarring in a rat model of myocardial infarction. Am J Physiol Heart Circ Physiol 2007; 292: H2809–H2816.
- 19.↑
Chetboul V, Sampedrano CC, Gouni V, et al. Ultrasonographic assessment of regional radial and longitudinal systolic function in healthy awake dogs. J Vet Intern Med 2006; 20: 885–893.
- 20.↑
Chetboul V, Gouni V, Sampedrano CC, et al. Assessment of regional systolic and diastolic myocardial function using tissue Doppler and strain imaging in dogs with dilated cardiomyopathy. J Vet Intern Med 2007; 21: 719–730.
- 21.↑
Chetboul V, Serres F, Gouni V, et al. Radial strain and strain rate by two-dimensional speckle tracking echocardiography and the tissue-based velocity technique in the dog. J Vet Cardiol 2007; 9: 69–81.
- 22.
Berthelsen PG, Eldrup N, Nilsson LB, et al. Thermodilution cardiac output. Cold vs room temperature injectate and the importance of measuring the injectate temperature in the right atrium. Acta Anaesthesiol Scand 2002; 46: 1103–1110.
- 23.
Moise SF, Sinclair CJ, Scott DH. Pulmonary artery blood temperature and the measurement of cardiac output by thermodilution. Anaesthesia 2002; 57: 562–566.
- 24.
Runciman WB, Ilsley AH, Roberts JG. An evaluation of thermodilution cardiac output measurement using the Swan-Ganz catheter. Anaesth Intensive Care 1981; 9: 208–220.
- 25.↑
Thomas WP, Gaber CE, Jacobs GJ, et al. Recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat. Echocardiography Committee of the Specialty of Cardiology, American College of Veterinary Internal Medicine. J Vet Intern Med 1993; 7: 247–252.
- 26.
Perk G, Tunick PA, Kronzon I. Non-Doppler two-dimensional strain imaging by echocardiography—from technical considerations to clinical applications. J Am Soc Echocardiogr 2007; 20: 234–243.
- 27.↑
Glantz SA, Slinker BK. Applied regression & analysis of variance. 2nd ed. New York: McGraw-Hill, 2001; 418–427.
- 28.↑
Warner RM. Applied statistics: from bivariate through multivariate techniques. Thousand Oaks, Calif: Sage Publications, 2008.
- 29.↑
Korinek J, Wang J, Sengupta PP, et al. Two-dimensional strain—a Doppler-independent ultrasound method for quantitation of regional deformation: validation in vitro and in vivo. J Am Soc Echocardiogr 2005; 18: 1247–1253.
- 30.
Korinek J, Kjaergaard J, Sengupta PP, et al. High spatial resolution speckle tracking improves accuracy of 2-dimensional strain measurements: an update on a new method in functional echocardiography. J Am Soc Echocardiogr 2007; 20: 165–170.
- 31.↑
Amundsen BH, Helle-Valle T, Edvardsen T, et al. Noninvasive myocardial strain measurement by speckle tracking echocardiography: validation against sonomicrometry and tagged magnetic resonance imaging. J Am Coll Cardiol 2006; 47: 789–793.
- 32.
Knebel F, Schattke S, Bondke H, et al. Evaluation of longitudinal and radial two-dimensional strain imaging versus Doppler tissue echocardiography in predicting long-term response to cardiac resynchronization therapy. J Am Soc Echocardiogr 2007; 20: 335–341.
- 33.
Arita T, Sorescu GP, Schuler BT, et al. Speckle-tracking strain echocardiography for detecting cardiac dyssynchrony in a canine model of dyssynchrony and heart failure. Am J Physiol Heart Circ Physiol 2007; 293: H735–H742.
- 34.
Suffoletto MS, Dohi K, Cannesson M, et al. Novel speckle-tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation 2006; 113: 960–968.
- 35.↑
Leitman M, Lysyansky P, Sidenko S, et al. Two-dimensional strain—a novel software for real-time quantitative echocardiographic assessment of myocardial function. J Am Soc Echocardiogr 2004; 17: 1021–1029.
- 36.
Yip G, Abraham T, Belohlavek M, et al. Clinical applications of strain rate imaging. J Am Soc Echocardiogr 2003; 16: 1334–1342.
- 37.
Marwick TH. Measurement of strain and strain rate by echocardiography: ready for prime time? J Am Coll Cardiol 2006; 47: 1313–1327.
- 38.↑
Bayram M, De Luca L, Massie MB, et al. Reassessment of dobutamine, dopamine, and milrinone in the management of acute heart failure syndromes. Am J Cardiol 2005; 96: 47G–58G.
- 39.
Becker M, Kramann R, Dohmen G, et al. Impact of left ventricular loading conditions on myocardial deformation parameters: analysis of early and late changes of myocardial deformation parameters after aortic valve replacement. Am Soc Echocardiogr 2007; 20: 681–689.
- 40.
Rösner A, Bijnens B, Hansen M, et al. Left ventricular size determines tissue Doppler-derived longitudinal strain and strain rate. Eur J Echocardiogr 2009; 10: 271–277.
- 41.
Urheim S, Edvardsen T, Torp H, et al. Myocardial strain by Doppler echocardiography. Circulation 2000; 102: 1158–1164.
- 42.↑
Lee JD, Tajimi T, Widmann TF, et al. Application of end-systolic pressure-volume and pressure-wall thickness relations in conscious dogs. J Am Coll Cardiol 1987; 9: 136–146.
