Congestive heart failure is a common clinical syndrome characterized, in part, by abnormal hemodynamics and foremost by increases in ventricular filling pressures in dogs1 and humans.2–4,a Development of pulmonary edema, a life-threatening event in left-sided CHF, is predicted largely by the magnitude of LVFP interacting with the diastolic properties of the left ventricle. Thus, LVFP is fundamental to the pathogenesis and clinical course of CHF and for the treatment of affected animals.1,5,6,a
A simple but quantifiable noninvasive method that predicts an increase in LVFP could promote the early recognition of CHF, guide optimal medical management, improve prognostication, and facilitate therapeutic monitoring. The LVFP or its surrogates (ie, mean LAP or LVEDP)7 can be measured directly, but cardiac catheterization is required, and this approach generally is unsuitable in clinical settings.1 Other indirect methods, such as physical examination,8,9 thoracic radiography,9–11 and measurement of circulating cardiac biomarkers (such as natriuretic peptides7,12,13), may also be used. However, such methods have unknown sensitivity and specificity, may require days for results, or may lack clinically useful discrimination limits.
Doppler echocardiography has emerged as the method of choice for noninvasive prediction of an increase in LVFP in humans.14–22 Several Doppler echocardiography variables based on transmitral flow,15,18,23–25 pulmonary venous flow,26–30 and tissue Doppler imaging16,17,21,25 have been proposed to estimate LVFP, and guidelines for their diagnostic use in human patients have been reported.16,17,22,24,31 However, similar investigations in dogs are rare32–37 and limited by the number of Doppler echocardiography variables evaluated. Therefore, far less is known about the applicability of Doppler echocardiography indices validated in human patients for the prediction of an increase in LVFP in dogs.
The study reported here was designed to validate and compare various Doppler echocardiography indices to invasively measure LAP in dogs. The study addressed the hypothesis that Doppler echocardiography can be used to predict an increase in LVFP in healthy anesthetized dogs subjected to volume loading. The main objective was to evaluate Doppler echocardiography as a predictor of LVFP over a range of hemodynamic values and heart rates that relate to compensated, untreated, and treated CHF.
Materials and Methods
Animals—Seven healthy purpose-bred hound dogs that weighed 17.1 to 25.0 kg (median, 21.6 kg) and were 1 to 3 years old were included in the study. Four dogs were females, and 3 were males. The study protocol was reviewed and approved by the Animal Care and Use Committee and the Institutional Review Board of the Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University. All animals were treated in compliance with the National Institutes of Health guidelines on the care and use of laboratory animals.
Anesthesia, instrumentation, and hemodynamic measurements—Each dog was sedated with acepromazine maleateb (0.025 mg/kg, IM) and butorphanol tartratec (0.25 mg/kg, IM). An 18-gauge over-the-needle catheter was placed in the right cephalic vein and in the right lateral saphenous vein, and a 20-gauge arterial over-the-needle catheter was placed in the right dorsal pedal artery (5 dogs) or right femoral artery (2 dogs). Anesthesia was induced approximately 30 minutes later by the administration of propofold (5 mg/kg, IV). After endotracheal intubation, anesthesia was maintained with isofluranee (0.5% to 3.0%) in 100% oxygen by use of mechanical ventilation with a tidal volume of 10 to 15 mL/kg and a respiratory frequency of 12 breaths/min.
Dogs were positioned in left lateral recumbency on a custom-made fluoroscopy table designed with a lateral aperture to allow continuous echocardiographic examinations with the transducer from underneath during cardiac catheterization and hemodynamic interventions. Drugs and fluids were infused into the right cephalic vein as boluses or as a constant rate infusion by use of a syringe pumpf and a high-rate infusion pump.g Cefazolinh (20 mg/kg, IV) was administered immediately before and 90 minutes and 8 hours after induction of anesthesia. Heparini (100 U/kg, IV) was given once after completion of instrumentation.
The right external jugular vein (all dogs) and right carotid artery (5 dogs) or left femoral artery (2 dogs) were surgically exposed. Lidocaine (2% solution)j was used to irrigate the surgical fields, reduce vessel spasm, and provide additional local anesthesia. A 9-F, 6-cm introducer sheathk was inserted into the artery, and fluoroscopic guidance was used to advance an 8-F, 100-cm high-fidelity dual manometer–tipped catheterl (with pressure transducers 4 cm apart) and a soft-tip, 150-cm, 0.032-inch guide wirem into the left ventricle. The catheter was positioned to simultaneously record left ventricular and aortic pressures. The micromanometer catheter was connectedn to a digital physiologic recording systemo–t via a control boxu for continual recording of left ventricular and central aortic pressures. A commercially available, 6-F, 44-cm transseptal introducer sheathv with a curved needlew (2 dogs) or a custom-made, stainless-steel, 10-F, 30-cm transseptal introducer sheath with a curved end and a plastic stylet (5 dogs) was used for transvenous atrial septotomy of the right external jugular vein under echocardiographic and fluoroscopic guidance. After penetration of the atrial septum in the area of the oval fossa, the stylet was removed, and proper position of the catheter was confirmed echocardiographically by use of contrast medium (5 mL of agitated saline [0.9% NaCl] solution administered via the introducer sheath into the left atrium). Fluoroscopic guidance was used to advance a 3-F, 100-cm, high-fidelity, dual micromanometer—tipped catheter (with pressure transducers 3 cm apart)x through the introducer sheath into the left atrium. The distal pressure transducer was positioned in a pulmonary vein, and the proximal transducer was positioned in the approximate center of the left atrium. The introducer sheath was then withdrawn, the micromanometer catheter secured, and the catheter connectedn to a digital physiologic recording systemo–t via a control boxu for continual recording of LAP. A 5-F, 80-cm, end-hole multipurpose cathetery and a 5-F, 80-cm, temporary pacing leadz were advanced via the right external jugular vein into the right atrium for continuous recording of RAP and intermittent right atrial pacing.aa The fluid-filled catheter was connected to a pressure transducerbb and a physiologic recording system.o–t The 3 catheters were sutured to the vein and surrounding tissues and were not moved during the experiments.
Transducers were balanced at atmospheric pressure, calibrated against a mercury manometer, and zero referenced at the level of the midventricle (left ventricular catheter) or midatrium (left atrial and right atrial catheters) immediately prior to use. Zero calibration of the right atrial catheter was achieved prior to each recording period. At the end of the experiments, the micromanometer-tipped catheters were placed in saline solution and exposed to air to confirm accurate registration of zero pressures.
The CO was determined by lithium dilution and arterial pulse contour analysis.cc,dd This method uses lithium dilution to determine CO and to calibrate a systemcc,dd that calculates CO continuously from the energy of the arterial pressure waveform. This method allows beat-to-beat determination of stroke volume and instantaneous determination of CO.38 The device has been validated in dogs, and measurements were performed as described elsewhere.39 Briefly, a small volume of lithium chlorideee was injected into the right cephalic vein via an injection kit,ff and arterial blood was moved past a lithium sensorgg to measure lithium concentration. The AUC for the resulting lithium dilution curve was used to derive the CO, which was then used to calibrate pulse contour analysis CO. In this experiment, the right dorsal pedal artery (5 dogs) or right femoral artery (2 dogs) was used for continuous pressure tracing and collection of arterial blood samples. Samples for blood gas analysis were obtained prior to each lithium dilution curve to assess total hemoglobin and serum sodium concentrations by use of a blood gas analyzer.hh The dose of lithium chloride for each curve was 0.15 mmol, which was administered into the right cephalic vein. Pulse contour analysis CO was used to monitor CO during the experiment, with a second lithium chloride calibration curve performed halfway through the volume loading period to verify accuracy.
Heart rate, body temperature, ECG, left ventricular pressure, LAP, RAP, aortic pressure, and CO were monitored continuously and recorded simultaneously during each treatment period. Body temperature was monitored by use of an esophageal thermometer positioned at the level of the heart base; body temperature was maintained by use of a circulating warm water blanket positioned beneath each dog.
At the conclusion of the study, the central catheters and peripheral arterial catheter were removed, the vessels were ligated, and the incision was closed in a routine mannerly. The neck was bandaged, and butorphanol (0.25 mg/kg, IV, q 8 h) was administered for 24 hours. The dogs were monitored in the intensive care unit for 12 hours and then returned to their ward. Amoxicillin-clavulanic acidii (20 mg/kg, IV, q 12 h) was administered for 5 days. Sutures were removed 12 days after the procedure, and all dogs were then offered for adoption to private homes.
Echocardiography—Transthoracic 2-dimensional spectral Doppler and tissue Doppler echocardiographic examinations were performed by a single investigator (KES). Each dog was positioned in left lateral recumbency, and echocardiographic examinations were performed by use of a digital ultrasonographic systemjj,kk with a 3.5-MHz transducer. Mechanical ventilation was stopped before each echocardiographic recording, and the time frames for the echocardiogram, physiologic hemodynamic monitoring system,o–t and CO metercc,dd were synchronized to facilitate simultaneous off-line data analysis. Sweep speed during recordings was 100 to 150 mm/s. Optimized left apical parasternal standard views of the left ventricular inflow and outflow tracts were used for data acquisition. The same inflow tract view was used to acquire images for evaluation of longitudinal motion of the lateral and septal mitral annulus. Two-dimensional cine loops and Doppler tracings were recorded and stored on the internal hard drive of the echocardiograph or on a magneto-optical disc and analyzed off-line. The system was set to store raw DICOM data. Simultaneously, a 1-lead ECG was recorded. Heart rate was calculated from the preceding R–R interval on the ECG. Measurements were obtained from digital still images as a mean of 5 to 12 consecutive cardiac cycles, which were obtained during end expiration. Only high-quality images were used for data analysis. All measurements were made off-line by 1 investigator (KES) who was unaware of the treatment stage and hemodynamic status.
Thirteen Doppler echocardiographic variables were measured and 8 Doppler echocardiographic variables were calculated as described in dogs40,41 and humans32,42 (Figure 1). In brief, by use of the left apical 3-chamber view, IVRT was measured as the period from the Doppler signal of aortic valve closure to the beginning of the E wave, with a 6- to 10-mm pulsed-wave sample volume placed in an intermediate position between the left ventricular inflow and outflow tracts. Transmitral flow was recorded by use of the left apical 2-chamber view, with the 2-mm pulsed-wave sample volume placed between the tips of the opened mitral valve leaflets. The A wave was then recorded. Peak velocities (peak E and peak A) were measured. When E and A waves were completely fused, peak velocity of the summated wave was measured. Partially fused E and A waves with a velocity > 20 cm/s at the point of intersection between the end of the E wave and beginning of the A wave (so-called E-at-A), which was evident in some dogs with NSR, were not used for data analysis. The DTE was measured as the interval from peak E to end of the E wave at the baseline. The Aduration was measured from the beginning to the end of the late diastolic inflow signal. Pulmonary venous flow was recorded by use of the same left apical 2-chamber view, with minimized baseline filter and with a 4- to 6-mm pulsed-wave sample volume placed 2 to 4 mm within the pulmonary vein of the left caudal lung lobe.41 Particular emphasis was placed on recording high-quality spectra of the AR wave. The peak AR and ARduration were measured. The AR wave was not analyzed in dogs with merged E and A waves. Dopplerderived velocities from myocardial motion were also recorded by use of an apical 2-chamber view, with a 5- to 7-mm sample volume placed in the septal or lateral corners of the mitral annulus. Frame rate during tissue Doppler studies was optimized (> 160 frames/s) by narrowing the tissue Doppler imaging sector. The Ea was assessed, and Ea sept, Ea lat, peak Ea sept, and peak Ea lat were measured. When the early and late components of mitral annular motion were partially fused, the wave forms were not measured. The intervals between the peak of the R wave and onset of the E wave and between peak of the R wave and onset of Ea sept and Ea lat were measured from nonsimultaneous images. For these measurements, only cardiac cycles with a cycle length within 5 milliseconds of that obtained for the R wave-to-E wave ratio25 were chosen.32 Subsequently, the TEa-E was calculated for both lateral and septal locations, and the mean value was derived.26 Color M-mode recordings of early diastolic left ventricular inflow were used to measure Vp and peak Vp. Color Doppler transmitral flow recordings were obtained, the Nyquist limit was reduced to approximately 50% of peak E, and slope of the first aliasing velocity line from the tip of the mitral valve to 3 cm into the left ventricular cavity was used for determination of Vp.43 Variables calculated included ratios for peak E:IVRT, peak E:peak A, Aduration:ARduration, peak E:peak Ea sept, peak E:peak Ea lat, peak E:peak Vp, and TEa-E.
Measurement reliability was determined for selected echocardiographic variables. Six echocardiograms from dogs with NSR or right atrial pacing were randomly selected from the pool of experiments to be subjected to 3 repeated analyses by 1 investigator (KES) to determine intraobserver measurement variability. The same 6 echocardiograms were subjected to independent repeated analyses by a second investigator (JAS) to determine interobserver measurement variability. Both investigators were unaware of the results of the prior echocardiographic analyses and the hemodynamic state.
Hemodynamic interventions—A wide range of hemodynamic states were induced by gradual volume loading with colloids,ll with both the RAP and mean LAP used to guide the administration of fluid volume. After baseline measurements, IV infusion of warmed hetastarch (100 mL/kg/h) was initiated. Mean LAP was monitored continuously, and hemodynamic and echocardiographic recordings were conducted simultaneously once target pressures of 10 to 15 mm Hg, 15 to 20 mm Hg, 20 to 25 mm Hg, 25 to 30 mm Hg, and 30 to 35 mm Hg were achieved. During Doppler echocardiographic recordings, fluid administration was temporarily reduced to maintain steady-state pressures. After the last target pressure was recorded, fluid administration was stopped, and furosemidemm (4 mg/kg, IV) was administered to decrease mean LAP to approximately 20 mm Hg. When LAP did not decrease substantially within 30 minutes after furosemide administration, a second dose of furosemide (2 mg/kg, IV) was administered (5 dogs). Hemodynamic and echocardiographic recordings were repeated when a mean LAP of approximately 20 mm Hg was achieved.
To study the effect of summated early and late diastolic transmitral flow signals on Doppler echocardiographic estimation of LVFP, the dogs also were subjected to right atrial pacing to cause fusion of the E and A waves. The lowest pacing rate needed to cause complete fusion was chosen (132 to 162 beats/min). Therefore, all recordings were made in duplicate at each target pressure stage with the dogs in NSR or in a paced rhythm.
Data analyses—Pressures, CO, and a single ECG lead were continuously recorded and stored digitally for subsequent analyses. Digital sampling rate for the micromanometer catheters was 500 samples/s (ie, 1 sample/2 ms). Pressures and computations were measured from the digital recordings at end expiration. Measurements included heart rate per minute, peak LAP at atrial contraction, peak LAP at ventricular contraction, LAP before atrial contraction, mean LAP (ie, mean pressure in the left atrium electronically derived from the AUC for LAP during 1 cardiac cycle), LVEDP, peak left ventricular systolic pressure, mean aortic pressure, mean RAP, peak rate of left ventricular pressure increase, and CO. The means of 5 consecutive measurements were calculated for CO or pressures. The LVEDP was defined as the left ventricular pressure immediately preceding the onset of ventricular contraction. Body surface area (in square meters) was calculated as follows44: (10.1 × [body weight in grams]0.67)/10,000. Cardiac index was calculated as CO divided by body surface area.
Statistical analysis—Statistical analyses were performed by use of commercially available software.nn,oo Observer variability was calculated by use of the following equation45: coefficient of variation = (mean difference between measurements/mean of measurements) × 100.
All data were visually inspected and tested for normality by use of the Kolmogorov-Smirnov test. Descriptive statistics were calculated for all invasive and Doppler echocardiographic variables. Differences between baseline values and values after hemodynamic interventions were examined by use of a 1-way repeatedmeasures ANOVA and Holm Sidak post hoc test.
Pooled data from all target pressures in all dogs were used to test the relationship between invasive and Doppler echocardiographic data. Pearson product moment correlation (for normally distributed residuals with constant variance) or Spearman rank order correlation (for nonnormally distributed residuals) and simple linear regression analysis were used to identify and quantify correlations between invasive and Doppler echocardiographic variables of LVFP. To correct for the influence of heart rate, IVRT and DTE were also indexed by dividing these variables by the square root of the duration of the preceding cycle,16 and analyses were then repeated. Scatterplots of predicted values for peak E:IVRT, IVRT, Aduration:ARduration, TEa-E, peak E:peak Ea, and Ea in relation to mean LAP were reported as means with 95% confidence intervals for the regression line and 95% confidence intervals for the entire population (prediction lines). An ROC curve analysis was used to determine optimal cutoff values for selected continuous Doppler echocardiographic variables for the prediction of an increase in LAP as defined by a mean LAP ≥ 15 mm Hg. Only Doppler echocardiographic variables with values of r > 0.70 as determined by prior correlation analysis were used for such procedures. The ROC represents the relationship between sensitivity and specificity, which is determined by plotting the rate for true-positive results (ie, sensitivity) against the rate for false-positive results (ie, specificity), because the cutoff point of the model varies. The AUC for the ROC curve was used as a summary measure for diagnostic accuracy and to quantify the ability of several continuous Doppler echocardiographic variables to predict an increase in LAP and was reported with 95% confidence intervals. For all analyses, values of P ≤ 0.05 were considered significant.
Results
Major complications were not evident during any of the experiments, and all dogs recovered from the procedures without complications. Minor complications included intermittent atrial ectopy (n = 4), ventricular ectopy (4), intermittent second-degree atrioventricular block during cardiac pacing (4), atrial flutter (2), slow ventricular tachycardia (1), mild pericardial effusion (1), clot formation at the tip of the left ventricular catheter (1), formation of a mass-like lesion at the atrial septotomy site (1), and moderate hemorrhage or oozing of blood with hematoma formation at the site of the incision in the neck (1). These abnormalities were transient and resolved after conventional treatments.
The total amount of fluid administered during high-rate volume infusion varied between 77 and 112 mL/kg over a time period of 52 to 105 minutes. Volume infusion led to a significant increase of LAP, with mean LAP varying from 2.4 to 34.0 mm Hg in dogs with NSR and 2.3 to 35.0 mm Hg in dogs with right atrial pacing (Table 1; Figure 2). Recordings of LAP and left ventricular pressure were obtained from dogs with NSR during the study (Figure 3). Mean LAP significantly (P < 0.001) correlated with peak LAP at atrial contraction, peak LAP at ventricular contraction, LAP before atrial contraction, and LVEDP in dogs with NSR (r = 0.95, 0.91, 0.97, and 0.95, respectively) and in dogs with right atrial pacing (r = 0.95, 0.99, 0.94, and 0.95, respectively). In dogs with NSR, heart rate varied from 71 to 161 beats/min and significantly increased with increasing LAP (r = 0.54; P < 0.001). In dogs with right atrial pacing, heart rate was between 132 and 162 beats/min and was significantly higher at all target pressures, compared with heart rate for dogs with NSR. Peak left ventricular systolic pressure varied from 84 to 121 mm Hg, mean aortic pressure varied from 65 to 121 mm Hg, mean RAP varied from 0.3 to 29.7 mm Hg, peak rate of left ventricular pressure increase varied from 1,263 to 2,779 mm Hg/s, and cardiac index varied from 1.6 to 9.4 L/min/m2.
Median (range) values for invasively measured indices of cardiac hemodynamics before (baseline) and during gradual volume loading to achieve target mean LAP in 7 clinically normal, anesthetized, volume-loaded dogs.
Variable | Cardiac rhythm | Mean LAP (mm Hg) | ||||||
---|---|---|---|---|---|---|---|---|
Baseline | 10–15 | 15–20 | 20–25 | 25–30 | 30–35 | 20* | ||
Heart rate (beats/min) | NSR | 106 (71–119) | 111 (96–120) | 109 (97–125) | 130 (120–158)† | 135 (125–161)† | 132 (125–146)† | 142 (128–157)† |
Paced | 140 (132–156)‡ | 140 (138–148)‡ | 145 (136–150)‡ | 150 (136–154)‡ | 160 (159–160)‡ | 161 (160–162)‡ | 160 (156–161)‡ | |
LAPa (mmHg) | NSR | 9.3 (4.8–14.5) | 15.9 (14.1–18.0)† | 19.0 (14.8–24.6)† | 25.9 (18.5–1.9)† | 30.8 (24.0–37.1)† | 36.6 (36.0–47.4)† | 23.5 (13.0–40.9)† |
Paced | 7.5 (3.5–14.9) | 17.1 (12.5–17.8)† | 23.2 (15.2–28.2)† | 27.6 (22.9–7.3)† | 28.1 (26.7–29.4)† | 32.6 (29.5–35.6)† | 21.2 (17.8–24.1)† | |
LAPv(mmHg) | NSR | 8.6 (4.6–15.8)† | 15.0 (12.4–17.9)† | 21.6 (13.3–23.0)† | 25.3 (14.0–9.7)† | 31.2 (21.6–33.8)† | 37.7 (35.1–38.1)† | 20.5 (13.3–28.4)† |
Paced | 9.2 (3.5–14.9) | 16.9 (15.2–18.4)† | 23.2 (12.9–26.6)† | 29.6 (18.5–2.0)†,‡ | 30.1 (24.8–35.3)† | 35.9 (29.1–42.7)† | 22.8 (16.5–27.0)† | |
LAPpre-A(mmHg) | NSR | 5.6 (1.9–11.9) | 9.7 (2.1–12.0) | 16.4 (13.5–19.2)† | 20.6 (15.7–24.9)† | 26.3 (17.6–28.4)† | 31.8 (28.5–34.1)† | 18.9 (11.0–27.7)† |
Paced | 5.8 (2.1–10.9) | 12.3 (1.8–16.5) | 18.2 (10.6–21.1)† | 25.1 (13.8–27.7)†,‡ | 24.4 (21.6–27.1)† | 28.6 (23.8–33.3)† | 19.3 (14.0–23.6)† | |
Mean LAP (mm Hg) | NSR | 7.2 (2.4–14.4) | 12.6 (9.8–15.3)† | 18.2 (11.3–20.7)† | 22.7 (16.1–25.5)† | 26.5 (19.0–31.6)† | 32.6 (31.6–34.0)† | 22.7 (10.4–24.6)† |
Paced | 7.1 (2.3–13.1) | 14.1 (11.4–16.0)† | 19.8 (10.3–22.0)† | 25.5 (16.1–27.3)† | 24.3 (21.4–27.1)† | 29.8 (24.6–35.0)† | 20.0 (14.7–23.3)† | |
LVEDP (mmHg) | NSR | 9.4 (4.0–12.6) | 15.2 (8.4–18.0)† | 18.0 (14.3–25.4)† | 24.8 (14.1–30.3)† | 30.0 (20.3–32.0)† | 34.5 (27.5–38.5)† | 20.9 (6.2–27.7)† |
Paced | 8.3 (2.1–12.1) | 17.4 (12.2–20.8)† | 21.3 (16.4–27.3)† | 26.0 (23.3–28.7)† | 27.5 (24.7–30.2)† | 32.0 (27.5–38.4)† | 19.6 (19.4–19.9)† | |
LVSP (mmHg) | NSR | 105 (84–120) | 97 (91–105) | 101 (91–123) | 105 (97–122) | 106 (91–139) | 108 (96–114) | 122 (96–136)† |
Paced | 101 (84–120) | 102 (95–111) | 105 (97–124) | 105 (97–132) | 101 (94–108) | 101 (96–106) | 116 (98–134)† | |
Mean AoP (mm Hg) | NSR | 91 (69–108) | 76 (67–100) | 89 (68–105) | 94 (75–105) | 95 (69–122) | 91 (78–102) | 109 (71–119)† |
Paced | 81 (65–105) | 82 (80–96) | 93 (76–104) | 93 (76–117) | 87 (74–100) | 88 (76–99) | 102 (78–121)† | |
Mean RAP (mm Hg) | NSR | 5.0 (1.5–10.0) | 12.2 (6.1–13.4)† | 16.3 (9.6–23.7)† | 17.1 (9.3–25.0)† | 20.7 (13.4–29.0)† | 25.7 (20.3–36.3)† | 14.6 (5.0–17.6)† |
Paced | 4.6 (0.3–10.2) | 12.3 (7.7–15.6)† | 17.6 (11.4–24.7)† | 20.4 (17.2–24.5)† | 21.1 (20.6–25.6)† | 26.3 (24.7–29.7)† | 15.9 (13.2–18.3)† | |
+dP/dtmax (mm Hg/s) | NSR | 1,934 | 1,601† | 1,467† | 1,550† | 1,635 | 1,530† | 1,983 |
(1,564–2,320) | (1,105–1,725) | (1,233–2,586) | (1,325–2,394) | (1,313–3,127) | (1,349–2,116) | (1,433–2,809) | ||
Paced | 1,837 | 1,763 | 1,601 | 1,688 | 1,562 | 1,597 | 1,924 | |
(1,385–2,308) | (1,263–1,975) | (1,361–2,625) | (1,320–2,779) | (1,313–1,810) | (1,401–1,593) | (1,438–2,409) | ||
Cardiac index (L/min/m2) | NSR | 2.8 (1.6–5.7) | 4.2 (3.0–5.2) | 3.7 (2.2–5.8) | 4.9 (2.7–7.4) | 5.3 (4.4–8.0)† | 5.3 (5.0–7.2)† | 5.3 (3.2–9.4)† |
Paced | 5.5 (2.3–8.5) | 6.1 (3.5–8.3) | 6.0 (2.6–7.3)‡ | 6.8 (3.0–7.3)† | 6.8 (5.2–8.4)† | 6.5 (5.3–7.1) | 5.0 (3.4–6.1) |
Target pressure was approximately 20 mm Hg.
Within a row, value differs significantly (P ≤ 0.05) from baseline value.
Within a variable within a column, value differs significantly (P ≤ 0.05) from value for NSR(pairwise comparison).
Paced = Right atrial pacing. LAPa = LAP at atrial contraction. LAPv = LAP at ventricular contraction. LAPpre-A = LAP immediately before left atrial contraction. LVSP = Left ventricular systolic pressure. AoP = Aortic pressure. +dP/dtmax = Peak rate of left ventricular pressure increase.
Twenty-nine to 49 paired Doppler echocardiographic–derived pressure measurements were available for comparative data analysis for various Doppler echocardiographic variables. Simultaneous echocardiographic data obtained during each treatment period were tabulated (Table 2). Increases in LAP led to an increase in peak E, peak A, peak E:IVRT, ARduration (only in dogs with NSR), peak E:peak Vp, and Ea and a decrease in IVRT, DTE, Aduration:ARduration (only in dogs with NSR), and TEa-E (only in dogs with right atrial pacing) but did not change peak E:peak A or peak E:peak Ea. Significant linear correlations between mean LAP and DE indices of LVFP were evident for all variables except peak E:peak A, peak Vp (only in dogs with right atrial pacing), and peak E:peak Ea (Table 3; Figures 4–6). Indexing IVRT and DTE to heart rate16 revealed significant (P < 0.001) but slightly different correlations to mean LAP than without indexing (r = −0.79 and r = −0.84 for IVRT divided by the square root of the duration of the preceding cycle in dogs with NSR or right atrial pacing, respectively, and r = −0.62 and r = −0.50 for DTE divided by the square root of the duration of the preceding cycle in dogs with NSR or right atrial pacing, respectively).
Mean (range) values for Doppler echocardiographic indices of left ventricular filling and LVFP before (baseline) and during gradual volume loading to achieve target mean LAP in 7 clinically normal, anesthetized, volume-loaded dogs.
Variable | Cardiac rhythm | Mean LAP (mm Hg) | ||||||
---|---|---|---|---|---|---|---|---|
Baseline | 10–15 | 15–20 | 20–25 | 25–30 | 30–35 | 20* | ||
Peak E(m/s) | NSR | 0.83 (0.53–1.08) | 0.75 (0.68–1.15) | 1.03 (0.79–1.23)† | 1.00 (0.91–1.21) | 1.09 (1.08–1.12)† | 1.17 (1.09–1.24)† | 1.02 (0.66–1.10) |
Paced | 0.76 (0.69–1.01) | 0.95 (0.76–1.03) | 1.04 (0.91–1.11)† | 1.15 (1.11–0.18)† | 1.27 (1.18–1.31)† | 1.23 (1.12–1.34)† | 0.99 (0.88–1.19) | |
Peak A(m/s) | NSR§ | 0.43 (0.38–0.52) | 0.48 (0.45–0.53) | 0.46 (0.35–0.51) | 0.59 (0.30–0.66)† | 0.58 (0.44–0.80)† | 0.67 (0.60–0.72)† | 0.53 (0.44–0.76) |
Paced | ND | ND | ND | ND | ND | ND | ND | |
Peak E: Peak A | NSR§ | 1.66 (1.23–2.84) | 1.64 (1.36–2.17) | 2.14 (1.59–2.91) | 1.81 (1.50–3.03) | 1.88 (1.25–2.73) | 1.74 (1.68–1.80) | 1.67 (1.45–2.30) |
Paced | ND | ND | ND | ND | ND | ND | ND | |
IVRT(ms) | NSR | 53 (43–62) | 51 (49–61) | 42 (37–54) | 34 (31–44)† | 33 (27–39)† | 28 (24–30)† | 43 (33–47)† |
Paced | 50 (46–67) | 46 (39–65) | 41 (34–46)† | 35 (28–41)† | 27 (25–28)†,‡ | 23 (20–26)†,‡42 (37–47)‡ | ||
Peak E:IVRT | NSR§ | 1.62 (1.12–1.80) | 1.36 (1.31–2.35) | 2.65 (1.55–2.84)† | 3.12 (2.06–3.90)† | 3.55 (2.85–4.30)† | 3.72 (3.57–5.20)† | 2.50 (2.12–3.06)† |
Paced | 1.73 (1.03–2.15) (4 65–4 92)† ‡ | 1.92 (1.57–2.26) | 2.63 (2.33–3.26)† | 3.05 (2.86–4.18)† | 4.79 | 5.48 (4.39–6.57)† | 2.86 (2.50–3.22)† | |
DTE(ms) | NSR§ | 100 (75–130) | 86 (44–103) | 77 (63–108) | 80 (52–89) | 68 (45–98)† | 61 (51–77)† | 91 (71–108) |
Paced | 81 (62–92)‡ | 84 (72–115) | 75 (60–91) | 68 (52–101) | 59 (55–61)† | 69 (66–72) | 85 (69–100) | |
Adurayion (ms) | NSR§ | 84 (62–93) | 96 (85–102) | 74 (71–88) | 83 (78–109) | 78 (67–108) | 70 (69–74) | 69 (63–83) |
Paced | ND | ND | ND | ND | ND | ND | ND | |
ARduration (ms) | NSR | 58 (56–79) | 71 (61–100) | 85 (53–102) | 88 (77–110)† | 82 (68–109)† | 98 (79–114)† | 82 (63–108)† |
Paced | ND | ND | ND | ND | ND | ND | ND | |
Aduration: | NSR§ | 1.33 (1.00–1.60) | 1.27 (0.97–1.67) | 1.12 (0.75–1.39) | 0.97 (0.71–1.35)† | 0.73 (0.65–1.00)† | 0.79 (0.65–0.87)† | 0.91 (0.61–1.32)† |
ARduration | Paced | ND | ND | ND | ND | ND | ND | ND |
Peak Vp(cm/s) | NSR∥ | 63.1 (50.9–77.9) | 50.9 (37.6–77.1) | 58.1 (35.6–80.1) | 50.4 (40.0–67.1) | 59.8 (20.0–94.1) | 54.9 (38.5–65.8) | 58.3 (19.6–73.5) |
Paced | 59.3 (47.4–71.0) | 44.7 (25.0–57.9)† | 75.3 (63.7–89.0) | 64.6 (27.9–70.2) | 50.7 (44.6–56.8) | 51.1 (32.1–70.2) | 55.5 (45.2–61.8) | |
Peak E:Peak Vp | NSR∥ | 1.09 (0.90–2.02) | 1.79 (0.88–2.13) | 1.86 (1.29–2.61)† | 1.81 (1.46–2.61)† | 1.86 (1.51–3.70)† | 2.07 (1.66–2.78)† | 1.60 (1.13–3.50) |
Paced | 1.34 (0.84–2.00) | 2.40 (1.31–3.52) | 1.30 (1.16–1.75)‡ | 1.81 (1.60–4.05)† | 3.73 (2.76–4.20)† | 2.70 (1.91–3.49)† | 1.80 (1.42–2.00) | |
Peak Ea sept (cm/s) | NSR | 6.8 (5.7–13.2) | 7.6 (6.3–8.0) | 8.7 (8.1–9.1) | 9.3 (8.3–12.2) | 12.4 (8.0–20.8)† | 10.9 (9.5–14.3) | 10.5 (7.9–13.1) |
Paced | 8.1 (5.3–9.4) | 8.8 (8.5–11.3) | 10.3 (9.5–10.7)† | 12.6 (9.9–3.7)†,† | 13.0 (10.5–15.5)† | 12.1 (9.5–14.7)† | 11.0 (9.4–12.1)† | |
Peak E: Peak Ea sept | NSR | 11.8 (4.0–15.1) | 10.5 (8.8–15.3) | 12.1 (8.5–13.6) | 9.8 (8.7–14.6) | 8.8 (6.8–14.0) | 10.7 (7.2–11.3) | 9.9 (5.1–11.8) |
Paced | 11.3 (7.3–17.2) | 9.7 (8.7–11.8) | 10.0 (8.7–11.7) | 9.1 (8.5–11.5) | 10.1 (8.4–11.8) | 10.9 (7.6–14.1) | 9.0 (7.8–9.9) | |
PeakEa la† (cm/s) | NSR | 9.3 (7.5–11.9) | 9.8 (7.2–12.2) | 10.7 (9.1–13.5) | 13.5 (9.2–14.6)† | 13.7 (11.0–19.7)† | 13.1 (12.5–13.9)† | 12.1 (10.3–14.2)† |
Paced | 9.1 (6.9–12.9) | 13.2 | 11.7 (10.5–13.1)† | 14.6 (11.0–19.3)† | 14.8 (13.8–15.9)† | 14.6 (14.2–15.0)† | 11.5 (9.9–12.4) | |
Peak E: PeakEa la† | NSR | 8.9 (6.5–10.8) | (11.9–14.5)†,†8.4 (6.4–11.1) | 10.6 (7.0–11.6) | 8.2 (6.7–10.7) | 7.9 (6.6–10.2) | 8.7 (7.4–9.1) | 8.4 (5.4–10.5) |
Paced | 8.8 (7.0–11.4) | 6.9 (6.4–7.7)† | 8.4 (7.9–10.6) | 7.7 (6.1–10.3) | 8.6 (7.8–9.5) | 8.5 (7.5–9.5) | 8.7 (7.3–9.7) | |
TE∂-E(ms) | NSR | 12 (−1 to 38) | 18 (−1 to 26) | 7 (−5 to 15) | 6 (−4 to 20) | 3 (−9 to 10) | 1 (−8 to 3) | 9 (−5 to 21) |
Paced | 16 (5 to 48) | 12 (−5 to 18) | 8 (−3 to 29) | 3 (−11 to 14)† | 1 (−1 to 4)† | −4 (−12 to 4)† | 7 (5 to 10)† |
Evaluations of 7 dogs for baseline and mean LAP of 10 to 15 mm Hg; 6 dogs for mean LAP of 15 to 20, 20 to 25, and 20 mm Hg; and 5 dogs for mean LAP of 25 to 30 and 30 to 35 mm Hg; flow patterns with fused E and A waves were discarded.
Evaluations of 5 dogs for baseline, and evaluations of 4 dogs for all other target pressures; flow patterns with poor quality were discarded.
ND = Not determined.
See Table 1 for remainder of key.
Coefficientsfor correlations between invasively derived variables and Doppler-derived estimates of LVFP.
Variable* | No. of observations | Cardiac rhythm | Mean LAP | LAPa | APv | LAPpre-a | LVEDP |
---|---|---|---|---|---|---|---|
Peak E (m/s) | 43 | NSR | 0.52 | 0.54 | 0.55 | 0.51 | 0.58 |
49 | Paced | 0.60 | 0.65 | 0.64 | 0.64 | 0.68 | |
Peak A (m/s) | 43 | NSR | 0.58 | 0.70 | 0.50 | 0.52 | 0.56 |
ND | Paced | ND | ND | ND | ND | ND | |
IVRT (ms) | 49 | NSR | −0.75 | −0.74 | −0.74 | −0.78 | −0.74 |
49 | Paced | −0.80 | −0.82 | −0.80 | −0.83 | −0.88 | |
Peak E:IVRT | 43 | NSR | 0.78 | 0.75 | 0.79 | 0.78 | 0.77 |
49 | Paced | 0.82 | 0.80 | 0.85 | 0.81 | 0.83 | |
DTE(ms) | 43 | NSR | −0.50 | −0.37 | −0.56 | −0.51 | −0.48 |
49 | Paced | −0.44 | NS | −0.45 | −0.44 | NS | |
Aduration (ms) | 43 | NSR | −0.58 | −0.59 | −0.60 | −0.61 | −0.47 |
ND | Paced | ND | ND | ND | ND | ND | |
ARduration (ms) | 43 | NSR | 0.60 | 0.61 | 0.59 | 0.59 | 0.59 |
ND | Paced | ND | ND | ND | ND | ND | |
Aduration:ARduration | 43 | NSR | −0.72 | −0.74 | −0.73 | −0.71 | −0.70 |
ND | Paced | ND | ND | ND | ND | ND | |
Peak Vp(cm/s) | 29 | NSR | −0.43 | −0.41 | −0.36 | NS | −0.40 |
ND | Paced | ND | ND | ND | ND | ND | |
Peak E: Peak Vp | 29 | NSR | 0.57 | 0.53 | 0.56 | 0.52 | 0.57 |
29 | Paced | 0.38 | 0.40 | 0.39 | NS | 0.43 | |
Peak Ea sept(cm/s) | 49 | NSR | 0.53 | 0.52 | 0.52 | 0.54 | 0.39 |
49 | Paced | 0.59 | 0.67 | 0.54 | 0.57 | 0.63 | |
Peak Ea lat(cm/s) | 49 | NSR | 0.58 | 0.52 | 0.60 | 0.62 | 0.49 |
49 | Paced | 0.63 | 0.67 | 0.61 | 0.60 | 0.64 | |
TEa-E(ms) | 49 | NSR | −0.46 | −0.44 | −0.40 | −0.44 | −0.47 |
49 | Paced | −0.71 | −0.73 | −0.68 | −0.66 | −0.69 |
Only variables with significant (P≤ 0.05) correlations are reported. NS = Correlation was not significant (P≥ 0.05). See Tables 1 and 2 for remainder of key.
The ROC curves used to predict a Doppler-derived mean LAP ≥ 15 mm Hg were similar for peak E:IVRT, IVRT, Aduration:ARduration, and TEa-E (only in dogs with right atrial pacing) with regard to shape, AUC, and 95% confidence interval (Figure 7). By use of ROC curve analysis, 3 diagnostic cutoff values for variables used for the diagnosis of an increase in LVFP (mean LAP ≥ 15 mm Hg) were identified, and the points of optimal test accuracy (as defined by the highest Youden index46 [ie, {sensitivity + specificity} − 1]) or data points with 100% sensitivity or 100% specificity were determined (Table 4). The value with the maximal Youden index was selected as a clinically relevant cutoff value to distinguish dogs with a typical LVFP from dogs with an increase in LVFP. The use of these cutoff values resulted in complete separation between groups for peak E:IVRT (only in dogs with right atrial pacing) and minimal overlap between groups for the other Doppler echocardiographic variables tested.
Sensitivity, specificity, and accuracy of various cutoff points for 4 Doppler echocardiographic variables of LVFP for use in the prediction of mean LAP ≥ 15 mm Hg.
Variable | Cardiac rhythm | Cutoff value | Sensitivity | Specificity | Accuracy* |
---|---|---|---|---|---|
Peak E:IVRT | NSR | 1.47 | 1.00 (0.88–1.00) | 0.50 (0.21–0.79) | 0.50 |
2.20† | 0.90 (0.74–0.98) | 0.92 (0.62–1.00) | 0.82 | ||
2.85 | 0.53 (0.34–0.72) | 1.00 (0.74–1.00) | 0.53 | ||
Paced | 2.17† | 1.00 (0.81–1.00) | 1.00 (0.63–1.00) | 1.00 | |
IVRT(ms) | NSR | 54 | 1.00 (0.88–1.00) | 0.31 (0.09–0.62) | 0.31 |
42† | 0.76 (0.57–0.90) | 0.93 (0.64–1.00) | 0.69 | ||
35 | 0.49 (0.30–0.68) | 1.00 (0.76–1.00) | 0.49 | ||
Paced | 47 | 1.00 (0.83–1.00) | 0.64 (0.31–0.90) | 0.64 | |
45† | 0.90 (0.67–0.99) | 0.91 (0.59–1.00) | 0.81 | ||
41 | 0.79 (0.55–0.94) | 1.00 (0.72–1.00) | 0.79 | ||
Aduration:ARduration | NSR | 1.43 | 1.00 (0.85–1.00) | 0.33 (0.09–0.66) | 0.33 |
1.04† | 0.83 (0.62–0.96) | 0.92 (0.62–1.00) | 0.75 | ||
1.00 | 0.70 (0.47–0.87) | 1.00 (0.74–1.00) | 0.70 | ||
Paced | ND | ND | ND | ND | |
TEa-E(ms) | NSR | ND | ND | ND | ND |
Paced | 21 | 1.00 (0.84–1.00) | 0.45 (0.14–0.79) | 0.45 | |
8† | 0.81 (0.58–0.95) | 0.89 (0.52–1.00) | 0.70 | ||
4 | 0.53 (0.30–0.75) | 1.00 (0.67–1.00 | 0.53 |
Values in parentheses are 95% confidence intervals.
Accuracy was calculated as follows: (sensitivity + specificity) −1.
Represents the cutoff value with the best combination for sensitivity and specificity.
See Tables 1 and 2 for remainder of key.
Results of the repeatability study were determined (Table 5). Variability in intraobserver measurements was less than variability in interobserver measurements. Except for peak Vp and TEa-E, right atrial pacing did not considerably affect coefficients of variation.
Intraobserver and interobserver coefficients of variation (CVs) for measured and calculated Doppler echocardiographic variables of LVFP from 6 randomly selected echocardiograms obtained from healthy anesthetized dogs.
Variable | Cardiac rhythm | Intraobserver CV (%) | Interobserver CV (%) |
---|---|---|---|
IVRT(ms) | NSR | 2.7 | 10.5 |
Paced | 2.1 | 12.5 | |
Peak E(m/s) | NSR | 0.7 | 1.7 |
Paced | 0.5 | 2.9 | |
Peak E:Peak A | NSR | 1.2 | 1.5 |
Paced | ND | ND | |
Peak E:IVRT | NSR | 3.0 | 13.3 |
Paced | 2.4 | 15.9 | |
DTE(ms) | NSR | 4.3 | 18.8 |
Paced | 2.6 | 9.4 | |
Aduration:ARduration | NSR | 4.1 | 16.5 |
Paced | ND | ND | |
Peak Vp (cm/s) | NSR | 4.5 | 20.6 |
Paced | 4.3 | 57.7 | |
Peak E:Peak Vp | NSR | 4.7 | 42.7 |
Paced | 2.8 | 43.9 | |
Peak Ea sept(cm/s) | NSR | 0.8 | 4.9 |
Paced | 0.6 | 5.1 | |
Peak Ea lat(cm/s) | NSR | 0.9 | 3.1 |
Paced | 0.4 | 3.3 | |
Peak E:Peak Ea sept | NSR | 1.0 | 2.4 |
Paced | 1.1 | 4.9 | |
Peak E:Peak Ea lat | NSR | 1.1 | 2.4 |
Paced | 2.9 | 4.0 | |
TEa-E(ms) | NSR | 11.4 | 26.7 |
Paced | 29.2 | 42.3 |
See Tables 1 and 2 for key.
Discussion
The study reported here has several important features, including that LAP as the criterion-referenced standard for assessment of LVFP7,47 was measured directly in all dogs by use of a high-fidelity micromanometer technique, preload was changed by controlled progressive volume loading with mean LAP reaching values compatible with those detected for severe CHF, merging of transmitral flow signals was induced by physiologic right atrial pacing as may result from sinus tachycardia, and a large number of Doppler echocardiographic variables commonly used to estimate LVFP were simultaneously acquired and compared with those obtained by an invasive standard. Moreover, 1 trained investigator, whose measurement variability was reported for all the assessed imaging variables, collected all echocardiographic imaging data.
Our results supported the hypothesis that Doppler echocardiographic variables, in particular peak E:IVRT, IVRT, Aduration:ARduration, and TEa-E, relate closely to mean LAP (a measure of LVFP) in dogs with induced acute volume overload in this study. On the contrary, conventional Doppler echocardiographic indices, including peak E, peak E:peak A, and DTE, or more recently suggested variables, such as peak E:peak Ea and peak E:peak Vp, were of limited value in the prediction of LAP. Moreover, an increase in heart rate associated with complete fusion of the Doppler transmitral inflow signals did not affect the diagnostic ability of the examined variables in the prediction of an increase in LAP. The latter finding is of particular clinical interest because merging of diastolic inflow signals is commonly evident in dogs with CHF and tachycardia attributable to increased sympathetic tone, which makes it impossible to estimate LVFP by use of traditional Doppler transmitral flow patterns.
In the study reported here, the relationship between peak E:IVRT and mean LAP was the strongest for the Doppler variables evaluated (accuracy of 82% and 100% in dogs with NSR and dogs with right atrial pacing, respectively) for the prediction of a mean LAP ≥ 15 mm Hg by use of the suggested cutoff values. This index has not been used commonly in the past for the Doppler echocardiographic prediction of LVFP, although a similar index, the (peak E:peak A)/IVRT ratio, can correlate closely (r = 0.93; P < 0.01)48 with pulmonary capillary wedge pressure (a surrogate measure of mean LAP)47 in humans with heart disease. The rationale behind the use of combined indices such as peak E:IVRT, as well as peak E:peak Vp, peak E:peak Ea, or peak E: strain rate, during isovolumic relaxation33 is to correct or adjust for the effect of relaxation on a variable that is largely dependent on filling pressure but that is also influenced by relaxation. By combining peak E, a variable that is determined mainly by LVFP and left ventricular relaxation,29,49–51 with a variable that is more dependent on relaxation,15,17,34,49,52,53 the effect of changes in relaxation on peak E can be minimized.20,21,23–25
Our study revealed only a modest correlation between LAP and peak E, with comparable results for dogs with NSR and right atrial pacing. This suggests that peak E cannot be simply correlated with a single hemodynamic factor, probably because of the multiple, interrelated factors that affect early diastolic filling. Similar results have been reported in humans, with coefficients of correlation between peak E and LVFP varying between 0.40 and 0.61.15,16,18,22,25,29,49 However, it is known from studies in laboratory dogs32,49,54 and humans15,19,21–23 that the accuracy of transmitral flow variables for the detection of left ventricular diastolic abnormalities or abnormal filling pressure is limited when left ventricular ejection fraction is preserved or there is a relaxation-dominant effect55 of left ventricular diastolic function as is evident with young animals or animals with myocardial ischemia or severe concentric left ventricular hypertrophy. Therefore, it is not surprising that, in our dogs with normal left ventricular systolic function, only modest associations between LAP and peak E and DTE, and no association between LAP and peak E:peak A, were detected.
The IVRT as an index of relaxation25,32,50,53,56,57 is linearly related to τ, the time constant of left ventricular isovolumic relaxation in cats,58 dogs,51,56 and humans,25,52 but it is also influenced by a multitude of other factors, including preload, afterload, heart rate, and age.40,50,59 Therefore, IVRT will represent the net effect of many determinants, of which relaxation is only one.52,59 Also, IVRT differs from τ in the sense that τ describes the rate of decline in left ventricular pressure during isovolumic relaxation, whereas IVRT measures the duration of the isovolumic relaxation period. Whereas a mild increase in left ventricular preload will shorten τ but does not influence IVRT,50 a moderate to severe increase in LVFP prolongs τ but shortens IVRT.51,52 This has been reported in experiments conducted in dogs53,54 and was confirmed in clinical studies 2,60 in humans with cardiomyopathy. A shortened IVRT is, by definition, an integral part of restrictive left ventricular filling,31,60 which is a flow pattern considered specific for advanced diastolic dysfunction, high LVFP, and CHF.2,60 High filling pressure may minimize the effect of relaxation on IVRT, which turns it into a more specific indicator of filling pressure. In the study reported here, IVRT was closely related to LAP, much more than to peak E, DTE, or peak E:peak A, which is a possible explanation as to why the ratio including IVRT (ie, peak E:IVRT) was superior to other variables for the prediction of mean LAP. Similar results have been reported by other investigators,16,18,19 although IVRT has rarely been used as a single surrogate of LVFP in clinical practice.25,48
It is known from studies in humans16,61 and dogs39 that heart rate affects IVRT and therefore its diagnostic performance. However, a mild to moderate increase in heart rate only minimally impacts IVRT,40 as was found in our study. Tachycardia accompanied by fusion of E and A waves improved the ability of IVRT to predict an increase in LVFP in the study reported here. Indexing of IVRT to the duration of the preceding cycle led to negligible improvement of correlation to mean LAP. Because both IVRT and peak E:IVRT had good predictive power, are easy to record, are simple to analyze, and are repeatable, both indices appear to be good candidates for use in clinical practice as Doppler echocardiographic variables of LVFP. However, studies in dogs with naturally developing heart disease are needed to substantiate such a recommendation.
When LVFP is increased, operating chamber compliance of the ventricle is reduced because it functions on a steeper portion of its pressure-volume curve. Working against a less compliant ventricle, less blood will enter the left ventricle, and there will be an increased pressure that builds up during late diastole. Thus, a greater force will retard the flow in the pulmonary veins, which increases the duration of retrograde flow (AR wave) from the left atrium into the pulmonary veins and reduces the duration of antegrade flow (A wave) from the left atrium into the left ventricle.26–28,35,62 The difference in the duration of flow with atrial contraction at the mitral valve and pulmonary veins and the ratio between both variables (Aduration:ARduration) can be used to identify an increase in LVFP in dogs35,37,62 and humans.14,21,22,25–28 In 1 study,27 investigators determined that when the pulmonary venous AR wave duration exceeds the duration of the mitral A wave, the LVEDP may be predicted to exceed 15 mm Hg with a sensitivity of 85% and specificity of 79% in humans with ischemic heart disease and cardiomyopathy. An inverse relationship between Aduration:ARduration and LVFP was reported.27,28 An r of −0.7027 and −0.7328 between LVFP and Aduration:ARduration is in agreement with our findings (r = −0.71; P < 0.001). Moreover, in a study14 of 176 consecutive cardiac patients that had undergone simultaneous Doppler echocardiography and invasive measurement of LVEDP, the Aduration:ARduration ratio was found to be the single best predictor of LVFP. The AUC for the prediction of LVEDP > 15 mm Hg was 0.91 for Aduration:ARduration,14 which is similar to the results of the study reported here. In contrast, peak E:peak Ea and peak E:peak Vp did not perform as well (AUC of 0.78 for peak E:peak Ea and 0.71 for peak E: peak Vp).14 Thus, the Aduration:ARduration ratio was useful in the estimation of LVFP in our study, and a ratio ≤ 1.0 predicted a mean LAP ≥ 15 mm Hg with 100% specificity. However, the practical use of the Aduration: ARduration ratio is hampered slightly by the fact that it cannot be used when transmitral flow waves overlap or when A or AR waves are missing, as in animals with sinus tachycardia, severe atrial mechanical dysfunction, or atrial fibrillation.26,35,37 In addition, it can be difficult to acquire high-quality pulmonary venous and transmitral flow recordings in some dogs.40,41
A novel index, the interval between the onset of the E and Ea waves (ie, TEa-E), has been reportedly related to left ventricular relaxation and LVFP in dogs32,63 and humans.25,32,64,pp The concept behind TEa-E is that in physiologic conditions, E and Ea waves are almost simultaneous but dissociate from each other with delayed relaxation or an increase in filling pressure. In situations in which filling pressure is high and left ventricular function is normal, TEa-E appears to be a specific indicator of filling pressure,64 whereas in situations characterized by left ventricular dysfunction with relaxation-dominant diastolic abnormalities, TEa-E more likely represents abnormal relaxation.32,63,64 In the study reported here, TEa-E was useful in the prediction of mean LAP, particularly during atrial pacing. For each increase or decrease of 1 mm Hg for mean LAP, TEa-E decreased or increased by approximately 1 millisecond, respectively. Paradoxically, the relationship between variables was inverse. Early transmitral flow began before mitral annulus motion in most dogs, and with increasing filling pressure, the difference between both processes decreased and even reversed at the highest mean LAP. The hemodynamic basis for such results cannot be fully explained, and our observation is in contrast to results in humans.32 A potential limitation of applying TEa-E clinically is the time-consuming process for its acquisition and its relatively poor repeatability. Also, R wave-to-Ea wave and R wave-to-E wave can only be measured by use of different cardiac cycles. Because heart rate influences systolic and diastolic time intervals,52 any difference in cycle duration between the cycles used to obtain measurements makes the use of TEa-E unreliable.25,32 The latter is the proposed reason that TEa-E predicted LVFP better in dogs with atrial pacing in which the R—R interval was constant throughout a given experiment. The clinical application of TEa-E in the evaluation of left ventricular diastolic function needs further investigation.
The peak E:peak Ea ratio and peak E:peak Vp ratio have been reported as useful Doppler echocardiographic indices of LVFP in numerous clinical studies in humans,6,14–20,22–24,32,34,65 with peak E:peak Ea preferred21,22 because of superior accuracy, ease of measurement, and better reproducibility.22 However, studies in dogs are rare, and to the authors' knowledge, only peak E:peak Ea has been validated.33,34,36 Both peak Ea23,34 and peak Vp20,43 are relatively preload-independent indices of left ventricular relaxation in situations in which left ventricular ejection fraction is low,22,33,65 which makes them suitable for correcting peak E for the effects of relaxation under such circumstances.15 However, results of the study reported here indicated that in healthy anesthetized dogs, peak Vp and, even more so, peak Ea are preload-dependent indices. Therefore, combined ratios between peak E:peak Vp and peak E:peak Ea were not useful in the prediction of mean LAP. This finding is in agreement with other reports from studies in healthy humans,22,23 healthy laboratory dogs undergoing open-chest surgery,34 or humans with cardiac disease and preserved left ventricular ejection fraction, which thereby limits the global use of such indices. In contrast, a close linear correlation was detected between mean LAP and peak E:peak Ea (r = 0.83; P < 0.05) by use of dogs with acute left ventricular volume overload secondary to severe experimentally induced mitral valve regurgitation.36 The differences between that study36 and results of the study reported here deserve discussion, although it is difficult to discern obvious reasons for these differences. Investigators in both studies used healthy dogs that were anesthetized, and the magnitude of left atrial hypertension induced was comparable. However, in that other study, integrity of the left ventricle was altered by experimental chordal rupture, which potentially affected filling dynamics. In addition, potent vasodilators were administered36 to induce additional hemodynamic changes, whereas dogs in our study were administered only fluids (in a controlled manner) to increase LAP. It is conceivable that such procedures36 may have induced load-independent changes that affected Doppler echocardiographic variables. Although preload significantly influenced peak E and peak Ea in both studies, such effects were less for peak E (increased by approx 40% in our study vs 100% during volume loading in that other study36) and more for peak Ea (increased by 40% to 60% in our study vs 30% during volume loading in that other study36). Conflicting results on the use of peak E: peak Ea as a reliable index of LVFP have been reported in humans with primary mitral valve regurgitation, with most studies25,66,67 rejecting the use of peak E:peak Ea under such circumstances. The strong preload dependency of Ea in hearts with preserved systolic function,25 as was found often in dogs with compensated degenerative mitral valve disease, may limit the use of peak E: peak Ea in the prediction of LVFP.25,66,67 However, peak E:peak Ea allows for an accurate estimation of LVFP in subjects with secondary mitral valve regurgitation and reduced systolic performance.67
We did not detect a consistent effect of LVFP and heart rate on Vp, which confirmed results obtained from other validation studies in dogs.43,68 Peak Vp was less dependent on preload than was peak Ea, and mean LAP and peak E:peak Vp were correlated. However, much closer correlation and more accurate predictive power of peak E:peak Vp have been reported in humans,14,15,22,23,43 particularly when left ventricular ejection fraction was normal.19,24 Because of the wide scatter of values within a certain range of measured LAPs, the fact that Vp is subject to considerable measurement error18,20,22 that results in poor repeatability, and the inferior performance in the accurate prediction of an increase in mean LAP compared with the accuracy for other Doppler echocardiographic variables, the clinical use of peak E:peak Vp in dogs cannot be recommended.
Because of the nature of the techniques used in our study, the associations detected and diagnostic cut-off values determined may not be valid for dogs with naturally developing chronic CHF and impaired left ventricular systolic function. Anesthesia may have affected cardiovascular function and thus interpretation of Doppler echocardiographic variables. Pooled data were used for regression analysis, and no attempts were made to determine the dog—target pressure–variable interaction by use of repeated-measures logistic regression. Progressive loading also significantly (P < 0.001) influenced heart rate. Therefore, the effect of changes in duration of the cardiac cycle could not be eliminated. The effect of dog on measurement variability was not specifically evaluated. Finally, most of the numerous Doppler echocardiographic methods applied to estimate LVFP are empiric because there is no physical relationship that can be used to extract absolute pressures from echocardiographic data. It is necessary to remember that Doppler echocardiographic variables measure blood flow and tissue motion, which are processes that are used only as indirect estimates of LVFP.
Overall, an increase in LVFP can be detected noninvasively by Doppler echocardiography in healthy anesthetized dogs with acute severe preload. The peak E:IVRT ratio was the single best predictor of LVFP, with adequate accuracy to discriminate dogs with a normal LVFP (mean LAP < 15 mm Hg) from dogs with an increased LVFP (mean LAP ≥ 15 mm Hg) to be used in isolation. However, whether Doppler echocardiography also provides objective noninvasive confirmation of the diagnosis of CHF and is useful in the quantification of LVFP in dogs with naturally developing heart disease remains to be determined.
ABBREVIATIONS
Aduration | Duration of the late diastolic transmitral flow wave |
ARduration | Duration of the late diastolic pulmonary vein atrial reversal wave |
AR wave | Late diastolic pulmonary vein atrial reversal wave |
AUC | Area under the curve |
A wave | Late diastolic transmitral flow wave |
CHF | Congestive heart failure |
CO | Cardiac output |
DICOM | Digital imaging and communications in medicine |
DTE | Deceleration time of the early diastolic transmitral flow wave |
Ea | Early diastolic motion wave of the mitral annulus |
Ea lat | Early diastolic motion wave at the lateral aspect of the mitral annulus |
Ea sept | Early diastolic motion wave of the septal aspect of the mitral annulus |
E wave | Early diastolic transmitral flow wave |
IVRT | Isovolumic relaxation time |
LAP | Left atrial pressure |
LVEDP | Left ventricular end-diastolic pressure |
LVFP | Left ventricular filling pressure |
NSR | Normal sinus rhythm |
Peak A | Peak velocity of the late diastolic transmitral flow wave |
Peak AR | Peak velocity of the late diastolic pulmonary vein atrial reversal wave |
Peak E | Peak velocity of the early diastolic transmitral flow wave |
Peak Ea lat | Peak velocity of early diastolic motion at the lateral aspect of the mitral annulus |
Peak Ea sept | Peak velocity of the early diastolic motion wave at the septal aspect of the mitral annulus |
Peak Vp | Peak velocity of early diastolic flow propagation |
RAP | Right atrial pressure |
ROC | Receiver operating characteristic |
TEa-E | Interval between the beginning of mitral annulus motion and the beginning of the early diastolic transmitral flow wave |
Vp | Velocity of early diastolic left ventricular flow propagation |
Liang HY, Cauduro S, Pellikka P, et al. Comparison of usefulness of newer echo-Doppler variables to left ventricular end-diastolic pressure in predicting heart failure (abstr). J Am Coll Cardiol 2006;47(suppl 2):100A–101A.
Acepromazine maleate injection, Boehringer Ingelheim Vetmedica Inc, St Joseph, Mo.
Butorphanol injection, IVX Animal Health Inc, St Joseph, Mo.
PropoFlo, Abbott Laboratories, North Chicago, Ill.
Isoflurane, Abbott Laboratories, North Chicago, Ill.
Medfusion 2010i syringe pump, Medexinc, Duluth, Ga.
Harvard apparatus peristaltic pump, model HA 66, Instech Laboratories Inc, Plymouth Meeting, Pa.
Cephazolin for injection, Sandoz Inc, Broomfield, Colo.
Heparin 1,000 U/mL, Abraxis Pharmaceuticals, Schaumburg, Ill.
Lidocaine HCl 2%, Abbott Laboratories, North Chicago, Ill.
9-F Check-Flo II introducer set, Cook Inc, Bloomington, Ind.
Millar dual-pressure transducer catheter, 8-F, model SPC-780C, Millar Instruments Inc, Houston, Tex.
GuideRight, 0.032 inch, 150 cm, St Jude Medical, Minnetonka, Minn.
Millar extension cable, model TEC-10C, Millar Instruments Inc, Houston, Tex.
MP100 workstation, BIOPAC Systems Inc, Santa Barbara, Calif.
AcqKnowledge, version 3.8.1 for Windows 2000, BioPac Systems, Santa Barbara, Calif.
DI-205, DATAQ Instruments, Akron, Ohio.
DI-720-USB, DATAQ Instruments, Akron, Ohio.
WINDAQ/PRO, DATAQ Instruments, Akron, Ohio.
ADVANCED CODAS, DATAQ Instruments, Akron, Ohio.
Millar control unit TCB-600, Millar Instruments Inc, Houston, Tex.
Mullins transseptal introducer sheath, Medtronic, Minneapolis, Minn.
Brockenbrough curved needle with a 19-gauge shaft and 21-gauge tip, Medtronic, Minneapolis, Minn.
Millar Micro-tip dual-pressure transducer catheter, model PC-751, Millar Instruments Inc, Houston, Tex.
5-F multipurpose catheter, 80 cm, Cook Inc, Bloomington, Ind.
Temporary pacing lead, Cordis Corp, Miami Lakes, Fla.
Medtronic 5375 demand pulse generator, Medtronic, Minneapolis, Minn.
Truewave disposal transducer, model TSD 104A, Edwards Lifesciences, Irvine, Calif.
LiDCO plus hemodynamic monitor, LiDCO Ltd, London, England.
LiDCO plus PulseCO screen, LiDCO Ltd, London, England.
CM40-044 lithium chloride ampoule, LiDCO Ltd, London, England.
CM50 injectate kit, LiDCO Ltd, London, England.
CM10 sensor, LiDCO Ltd, London, England.
ABL 725, Radiometer America, Westlake, Ohio.
Clavamox 375-mg tablets, Pfizer Animal Health, Exton, Pa.
Vivid 7 Vantage, GE Medical Systems, Milwaukee, Wis.
EchoPac software package, version BT04, GE Medical Systems, Milwaukee, Wis.
6% hetastarch in 0.9% sodium chloride, Hospira Inc, Lake Forest, Ill.
Furosemide injection, IVX Animal Health Inc, St Joseph, Mo.
Sigma Stat, version 3.5, SPSS Inc, Chicago, Ill.
Prism 4, Graph Pad Software Inc, San Diego, Calif.
Min PK, Ha JW, Jung JH, et al. Value of measuring the time difference between onset of mitral inflow and onset of early diastolic mitral annulus velocity for the evaluation of left ventricular diastolic pressures in patients with normal systolic function and intermediate E/E' (abstr). J Am Coll Cardiol 2006;47(suppl 2):96A–98A.
References
- 1.↑
Lord PF. Left ventricular diastolic stiffness in dogs with congestive cardiomyopathy and volume overload. Am J Vet Res 1976;37:953–957.
- 2.↑
St Goar FG, Masuyama T, Alderman EL. Left ventricular diastolic dysfunction in end-stage dilated cardiomyopathy: simultaneous Doppler echocardiography and hemodynamic evaluation. J Am Soc Echocardiogr 1991;4:349–360.
- 3.
Dougherty AH, Naccarelli GV, Gray EL. Congestive heart failure with normal systolic function. Am J Cardiol 1984;54:778–782.
- 4.
Pinamonti B, Di Lenard. A, Sinagra G. Restrictive left ventricular filling pattern in dilated cardiomyopathy assessed by Doppler echocardiography: clinical, echocardiographic and hemodynamic correlations and prognostic implications. Heart Muscle Disease Study Group. J Am Coll Cardiol 1993;22:808–815.
- 5.
Stevenson LW. Are hemodynamic goals viable in tailoring heart failure therapy? Circulation 2006;113:1020–1033.
- 6.
Dokainish H, Zoghbi WA, Lakkis NM, et al. Incremental predictive power of B-type natriuretic peptide and tissue Doppler echocardiography in the prognosis of patients with congestive heart failure. J Am Coll Cardiol 2005;45:1223–1226.
- 7.↑
Geske JB, Sorajja P, Nishimura RA, et al. Evaluation of left ventricular filling pressures by Doppler echocardiography in patients with hypertrophic cardiomyopathy. Correlation with direct left atrial pressure measurement at cardiac catheterization. Circulation 2007;116:2702–2708.
- 8.
Collins SP, Lindsell CJ, Peacock WF, et al. The combined utility of an S3 heart sound and B-type natriuretic peptide levels in emergency department patients with dyspnea. J Card Fail 2006;12:286–292.
- 9.
Chakko S, Woska D, Martinez H, et al. Clinical, radiographic, and hemodynamic correlations in chronic congestive heart failure: conflicting results may lead to inappropriate care. Am J Med 1991;90:353–359.
- 10.
Balbarini A, Limbruno U, Bertoli D, et al. Evaluation of pulmonary vascular pressures in cardiac patients: the role of the chest roentgenogram. J Thorac Imaging 1991;6:62–68.
- 11.
Henriksson L, Sundin A, Smedby O, et al. Assessment of congestive heart failure in chest radiographs. Observer performance with two common film-screen systems. Acta Radiol 1990;31:469–471.
- 12.
Cioffi G, Tarantini L, Stefenelli C, et al. Changes in plasma N-terminal proBNP levels and ventricular filling pressures during intensive unloading therapy in elderly with decompensated congestive heart failure and preserved left ventricular systolic function. J Card Fail 2006;12:608–615.
- 13.
McCullough PA, Nowak RM, McCord J, et al. B-type natriuretic peptide in clinical judgment in emergency diagnosis of heart failure. Analysis from Breathing Not Properly (BNP) multinational study. Circulation 2002;106:416–422.
- 14.↑
Poerner TC, Goebel B, Kralev S, et al. Impact of mitral E/A ratio on the accuracy of different echocardiographic indices to estimate left ventricular end-diastolic pressure. Ultrasound Med Biol 2007;33:699–707.
- 15.↑
Schwammenthal E, Popescu BA, Popescu AC, et al. Association of left ventricular filling parameters assessed by pulsed wave Doppler and color M-mode Doppler echocardiography with left ventricular pathology, pulmonary congestion, and left ventricular end-diastolic pressure. Am J Cardiol 2004;94:488–491.
- 16.↑
Nagueh SF, Mikati I, Kopelen HA, et al. Doppler estimation of left ventricular filling pressure in sinus tachycardia. A new application of tissue Doppler imaging. Circulation 1998;98:1644–1650.
- 17.
Nagueh SF, Middleton KJ, Kopelen HA, et al. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 1997;30:1527–1533.
- 18.
Nagueh SF, Kopelen HA, Quinones MA. Assessment of left ventricular filling pressures by Doppler in the presence of atrial fibrillation. Circulation 1996;94:2138–2145.
- 19.
González-Vilchez F, Ayuela J, Ares M, et al. Comparison of Doppler echocardiography, color M-mode Doppler, and Doppler tissue imaging for the estimation of pulmonary capillary wedge pressure. J Am Soc Echocardiogr 2002;15:1245–1250.
- 20.
Garcia MJ, Ares MA, Asher C, et al. An index of early left ventricular filling that combined with pulsed Doppler peak E velocity may estimate capillary wedge pressure. J Am Coll Cardiol 1997;29:448–454.
- 21.
Ommen SR, Nishimura RA, Appleton CP, et al. Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures. A comparative simultaneous Doppler-catheterization study. Circulation 2000;102:1788–1794.
- 22.↑
Kidawa M, Coignard L, Drobinski G, et al. Comparative value of tissue Doppler imaging and M-mode color Doppler mitral flow propagation velocity for the evaluation of left ventricular filling pressure. Chest 2005;128:2544–2550.
- 23.
Firstenberg MS, Levine BD, Garcia MJ, et al. Relationship of echocardiographic indices to pulmonary capillary wedge pressures in healthy volunteers. J Am Coll Cardiol 2000;36:1664–1669.
- 24.
Su HM, Lin TH, Voon WC, et al. Determination of pulmonary capillary wedge pressure using pulsed wave Doppler echocardiography: clinical application of range ambiguity phenomenon. J Am Soc Echocardiogr 2005;18:1023–1029.
- 25.↑
Diwan A, McCulloch M, Lawrie GM, et al. Doppler estimation of left ventricular filling pressures in patients with mitral valve disease. Circulation 2005;111:3281–3289.
- 26.↑
Sohn DW, Choi YJ, Oh BH, et al. Estimation of left ventricular end-diastolic pressure with the difference in pulmonary venous and mitral A durations is limited when mitral E and A waves are overlapped. J Am Soc Echocardiogr 1999;12:106–112.
- 27.↑
Rossvoll O, Little LK. Pulmonary venous flow velocities recorded by transthoracic Doppler ultrasound: relation to left ventricular diastolic pressures. J Am Coll Cardiol 1993;21:1687–1696.
- 28.↑
Yamamoto K, Nishimura RA, Burnett JC, et al. Assessment of left ventricular end-diastolic pressure by Doppler echocardiography: contribution of duration of pulmonary venous versus mitral flow velocity curves at atrial contraction. J Am Soc Echocardiogr 1997;10:52–59.
- 29.
Kuecherer HF, Muhiudeen IA, Kusumoto FM, et al. Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow. Circulation 1990;82:1127–1139.
- 30.
Tabata T, Thomas JD, Klein AL. Pulmonary venous flow by Doppler echocardiography: revisited 12 years later. J Am Coll Cardiol 2003;41:1243–1250.
- 31.
Thomas JD, Popovic ZB. Assessment of left ventricular function by cardiac ultrasound. J Am Coll Cardiol 2006;48:2012–2025.
- 32.↑
Rivas-Gotz C, Khoury DS, Manolios M, et al. Time interval between onset of mitral inflow and onset of early diastolic velocity by tissue Doppler: a novel index of left ventricular relaxation. J Am Coll Cardiol 2003;42:1463–1470.
- 33.↑
Wang J, Khoury DS, Thohan V, et al. Global diastolic strain rate for the assessment of left ventricular relaxation and filling pressures. Circulation 2007;115:1376–1383.
- 34.↑
Jacques DC, Pinsky MR, Severyn D, et al. Influence of alterations in loading on mitral annular velocity by tissue Doppler echocardiography and its associated ability to predict filling pressures. Chest 2004;126:1910–1918.
- 35.
Appleton CP. Hemodynamic determinants of Doppler pulmonary venous flow velocity components: new insights from studies in lightly sedated normal dogs. J Am Coll Cardiol 1997;30:1562–1574.
- 36.↑
Oyama MA, Sisson DD, Bulmer BJ, et al. Echocardiographic estimation of mean left atrial pressure in a canine model of acute mitral valve insufficiency. J Vet Intern Med 2004;18:667–672.
- 37.
Akita S, Ohte N, Hashimoto T, et al. Effects of volume loading on pulmonary venous flow patterns in dogs with normal left ventricular function. Angiology 1995;46:393–399.
- 38.↑
Hamilton TT, Huber LM, Jessen ME. PulseCo: a less-invasive method to monitor cardiac output from arterial pressure after cardiac surgery. Ann Thorac Surg 2002;74:S1408–S1412.
- 39.↑
Mason DJ, O'Grady M, Woods JP, et al. Assessment of lithium dilution cardiac output as a technique for measurement of cardiac output in dogs (Erratum published in Am J Vet Res 2001;62:1611). Am J Vet Res 2001;62:1255–1261.
- 40.↑
Schober KE, Fuentes VL. Effects of age, body weight, and heart rate on transmitral and pulmonary venous flow in clinically normal dogs. Am J Vet Res 2001;62:1447–1454.
- 41.↑
Schober KE, Luis Fuente V, McEwan JD, et al. Pulmonary venous flow characteristics as assessed by transthoracic pulsed Doppler echocardiography in normal dogs. Vet Radiol Ultrasound 1998;39:33–41.
- 42.
Appleton CP, Firstenberg MS, Garcia MJ, et al. The echo-Doppler evaluation of left ventricular diastolic function. A current perspective. Cardiol Clin 2000;18:513–546.
- 43.↑
Garcia MJ, Smedira NG, Greenberg NL, et al. Color M-mode Doppler flow propagation velocity is a preload insensitive index of left ventricular relaxation: animal and human validation. J Am Coll Cardiol 2000;35:201–208.
- 44.↑
Theilen GH. Chapter 6: veterinary medical oncology. In: Ettinger SJ, ed. Textbook of veterinary internal medicine: diseases of the dog and cat. Philadelphia: WB Saunders Co, 1975;127–149.
- 45.↑
Atkinson G, Nevill AM. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med 1998;26:217–238.
- 46.↑
Taube A. Sensitivity, specificity, and predictive values: a graphical approach. Stat Med 1986;5:585–591.
- 47.↑
Chaliki HP, Hurrell DG, Nishimura RA, et al. Pulmonary venous pressure: relationship to pulmonary artery, pulmonary wedge, and left atrial pressure in normal, lightly sedated dogs. Catheter Cardiovasc Interv 2002;56:432–438.
- 48.↑
Berger M, Bach M, Hecht S, et al. Estimation of pulmonary artery wedge pressure by pulsed Doppler echocardiography and phonocardiography. Am J Cardiol 1992;69:562–564.
- 49.
Choong CY, Abascal VM, Thomas J, et al. Combined influence of ventricular loading and relaxation on the transmitral flow velocity profile in dogs measured by Doppler echocardiography. Circulation 1988;78:672–683.
- 50.↑
Fragata J, Arreias JC. Effects of gradual volume loading on left ventricular diastolic function in dogs: implications for the optimization of cardiac output. Heart 1996;75:352–357.
- 51.
Nishimura RA, Abel MD, Hatle LK, et al. Significance of Doppler indices of diastolic filling of the left ventricle: comparison with invasive hemodynamics in a canine model. Am Heart J 1989;118:1248–1258.
- 52.↑
Gamble WH, Shaver JA, Alvares RF, et al. A critical appraisal of diastolic time intervals as a measure of relaxation in left ventricular hypertrophy. Circulation 1983;68:76–87.
- 53.
Raff GL, Glantz SA. Volume loading slows left ventricular isovolumic relaxation rate. Evidence of load-dependent relaxation in the intact dog heart. Circ Res 1981;48:813–824.
- 54.
Yellin EL. Determinants of transmitral flow patterns: insights from the experimental laboratory. Am J Card Imaging 1990;4:147–154.
- 55.↑
Nishimura RA, Appleton CP, Redfield MM, et al. Noninvasive Doppler echocardiographic evaluation of left ventricular filling pressures in patients with cardiomyopathies: a simultaneous Doppler echocardiographic and cardiac catheterization study. J Am Coll Cardiol 1996;28:1226–1233.
- 56.
Yellin EL, Hori M, Yoran C, et al. Left ventricular relaxation in the filling and nonfilling intact canine heart. Am J Physiol Heart Circ Physiol 1986;250:H620–H629.
- 57.
Weiss JL, Frederiksen JW, Weisfeldt ML. Hemodynamic determinants of the time-course of fall in canine left ventricular pressure. J Clin Invest 1976;58:751–760.
- 58.↑
Schober KE, Fuentes VL, Bonagura JD. Comparison between invasive hemodynamic measurements and noninvasive assessment of left ventricular diastolic function by use of Doppler echocardiography in healthy anesthetized cats. Am J Vet Res 2003;64:93–103.
- 59.
Banerjee A. Impaired left ventricular relaxation is an early manifestation of diastolic dysfunction: can non-invasive indices be of help? Prog Pediatr Cardiol 1999;10:65–74.
- 60.
Appleton CP, Hatle LK, Popp RL. Relation of transmitral flow velocity patterns to left ventricular diastolic function: new insights from a combined hemodynamic and Doppler echocardiographic study. J Am Coll Cardiol 1988;12:426–440.
- 61.
Holmgreen SM, Goldberg SJ, Donnerstein RL. Influence of age, body size, and heart rate on left ventricular diastolic indexes in young subjects. J Am Coll Cardiol 1991;68:1245–1247.
- 62.
Steen T, Voss BMR, Smiseth OA. Influence of heart rate and left atrial pressure on pulmonary venous flow patterns in dogs. Am J Physiol Heart Circ Physiol 1994;266:H2296–H2302.
- 63.
Hasegawa H, Little WC, Ohno M, et al. Diastolic mitral annular velocity during the development of heart failure. J Am Coll Cardiol 2003;41:1590–1597.
- 64.↑
Choi JO, Park SW, Shin DH, et al. Preload dependency of the time interval between the onset of mitral inflow and the early diastolic mitral annular motion: a hemodialysis-related preload reduction study. Circ J 2007;71:669–674.
- 65.
Hsiao SH, Huang WC, Sy LC, et al. Doppler tissue imaging and color M-mode flow propagation velocity: are they really preload independent? J Am Soc Echocardiogr 2005;18:1277–1284.
- 66.
Bruch C, Stypmann J, Gradaus R, et al. Usefulness of tissue Doppler imaging for estimation of filling pressures in patients with primary and secondary pure mitral regurgitation. Am J Cardiol 2004;93:324–328.
- 67.↑
Olson JJ, Costa SP, Young CE, et al. Early mitral filling/diastolic mitral annular velocity ratio is not a reliable predictor of left ventricular filling pressure in the setting of severe mitral regurgitation. J Am Soc Echocardiogr 2006;19:83–87.
- 68.
Stugaard M, Smiseth OA, Risoe C, et al. Intraventricular early diastolic filling during acute myocardial ischemia in dogs: assessment by multigated color M-mode Doppler. Circulation 1993;88:2705–2713.