Author + information
- Received November 10, 1997
- Revision received March 12, 1998
- Accepted April 8, 1998
- Published online July 1, 1998.
- Paul A Grayburn, MD, FACCa,* (, )
- James L Weiss, MD, FACC∗,
- Terrence C Hack, MD, FACC†,
- Elizabeth Klodas, MD, FACC∗,
- Joel S Raichlen, MD, FACC‡,
- Manni A Vannan, MD, FACC§,
- Allan L Klein, MD, FACC∥,
- Dalane W Kitzman, MD, FACC¶,
- Steven G Chrysant, MD, FACC#,
- Jerald L Cohen, MD, FACC∗∗,
- David Abrahamson, MD, FACC††,
- Elyse Foster, MD, FACC‡‡,
- Julio E Perez, MD, FACC§§,
- Gerard P Aurigemma, MD, FACC∥,
- Julio A Panza, MD, FACC¶,
- Michael H Picard, MD, FACC##,
- Benjamin F Byrd III, MD, FACC∗∗∗,
- Douglas S Segar, MD, FACC†††,
- Stuart A Jacobson, MD, FACC‡‡,
- David J Sahn, MD, FACC§§ and
- Anthony N DeMaria, MD, FACC∥
- ↵*Address for correspondence: Dr. Paul A. Grayburn, Echocardiography Laboratories, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, Texas, 75216-9047
Objectives. This study was performed to compare the safety and efficacy of intravenous 2% dodecafluoropentane (DDFP) emulsion (EchoGen) with that of active control (sonicated human albumin [Albunex]) for left ventricular (LV) cavity opacification in adult patients with a suboptimal echocardiogram.
Background. The development of new fluorocarbon-based echocardiographic contrast agents such as DDFP has allowed opacification of the left ventricle after peripheral venous injection. We hypothesized that DDFP was clinically superior to the Food and Drug Administration–approved active control.
Methods. This was a Phase III, multicenter, single-blind, active controlled trial. Sequential intravenous injections of active control and DDFP were given 30 min apart to 254 patients with a suboptimal echocardiogram, defined as one in which the endocardial borders were not visible in at least two segments in either the apical two- or four-chamber views. Studies were interpreted in blinded manner by two readers and the investigators.
Results. Full or intermediate LV cavity opacification was more frequently observed after DDFP than after active control (78% vs. 31% for reader A; 69% vs. 34% for reader B; 83% vs. 55% for the investigators, p < 0.0001). LV cavity opacification scores were higher with DDFP (2.0 to 2.5 vs. 1.1 to 1.5, p < 0.0001). Endocardial border delineation was improved by DDFP in 88% of patients versus 45% with active control (p < 0.001). Similar improvement was seen for duration of contrast effect, salvage of suboptimal echocardiograms, diagnostic confidence and potential to affect patient management. There was no difference between agents in the number of patients with adverse events attributed to the test agent (9% for DDFP vs. 6% for active control, p = 0.92).
Conclusions. This Phase III multicenter trial demonstrates that DDFP is superior to sonicated human albumin for LV cavity opacification, endocardial border definition, duration of effect, salvage of suboptimal echocardiograms, diagnostic confidence and potential to influence patient management. The two agents had similar safety profiles.
Echocardiography is the most widely used cardiac imaging modality in the United States, with >5 million studies performed annually (1). Its most common clinical indication is to assess global and regional left ventricular (LV) function (2), a task that depends on the ability to distinguish the endocardial border from the LV blood pool. Unfortunately, endocardial border definition is often limited by poor acoustic windows or endocardial “dropout” in the apical views (3,4). This problem has fueled efforts to develop transpulmonary contrast agents that can opacify the LV cavity after intravenous injection, thereby accurately identifying the endocardial surface and enhancing the ability to measure LV size and function (5,6).
The first transpulmonary agent to be Food and Drug Administration (FDA)–approved for LV opacification was Albunex, air-filled microbubbles of sonicated 5% human albumin (7–9). Despite its early promise, this agent has not been widely used in echocardiography, in part because the air-filled microbubbles often do not persist in the circulation long enough for adequate imaging. Newer contrast agents use fluorocarbon gases, which, owing to their low diffusivity and blood solubility, persist longer in the circulation (10). A 2% dodecafluoropentane (DDFP) emulsion (EchoGen) has been developed as a novel phase-shift colloidal contrast agent containing liquid droplets of DDFP that convert to gaseous microbubbles after hypobaric activation (11). After intravenous administration, the microbubbles rapidly distribute in the vascular space (distribution half-life <90 s) and are subsequently eliminated from the lungs, without metabolism, through expired air. The half-life of DDFP is 2 min in blood and 10 to 16 min in expired air (12).
The present study was performed to compare the efficacy of intravenous DDFP with that of active control (Albunex) for LV cavity opacification in adult patients with a suboptimal echocardiogram. A secondary objective was to compare the clinical utility of the two agents for facilitating diagnosis and management. Finally, we compared the safety of these two echocardiographic contrast agents.
This Phase III, multicenter, single-blind, active-controlled trial was performed at 18 centers in the United States. The active control agent used in this study was sonicated 5% human albumin (Albunex, Mallinckrodt Medical), an FDA-approved device for LV opacification during echocardiography (9). To be enrolled in the study, patients had to have a suboptimal echocardiogram, defined as one in which the endocardial borders were not visible in at least two segments in either the apical two- or four-chamber views. Exclusion criteria included 1) known history of anaphylaxis or drug allergy requiring treatment; 2) premenopausal patients who were lactating, pregnant or potentially pregnant; 3) any investigational drug within 7 days; 4) history of alcohol or drug abuse; 5) clinically significant pulmonary disease manifested by active wheezing, hypoxia or breathlessness; 6) history of sleep apnea and obesity (>130% of ideal body weight); 7) serious or life-threatening condition; 8) New York Heart Association functional class III or IV heart failure; 9) unstable neurologic disease; 10) known or suspected right to left shunt; 11) unstable angina; 12) recent myocardial infarction (<6 months); or 13) serious cardiac arrhythmias.
Written informed consent was obtained in all subjects. A 20-gauge intravenous catheter was placed in the right antecubital fossa, and the patient was positioned in the left lateral decubitus position. Vital signs, pulse oximetry and a 12-lead electrocardiogram (ECG) were recorded at baseline. A blood sample was obtained and sent to a core laboratory for complete blood count, serum electrolytes, liver function tests and renal function tests. Baseline echocardiography was performed using commercially available equipment operating at a transmit and receive frequency of 2.5 MHz. Instrument settings, including gain, depth and gray-level compression, were optimized for each patient and then held constant for all baseline and postcontrast images. Apical two- and four-chamber views were required to determine efficacy; other views were optional. All studies were recorded on ½-in. SVHS videotape for subsequent analysis.
After the baseline echocardiogram was recorded, the active control agent was administered as a bolus injection of 0.22 ml/kg. Repeat echocardiographic images were recorded during the injection and until contrast effect was no longer visible in the left ventricle. Blood pressure, heart rate, respiratory rate and oximetry were measured eight times during the 15 min after injection. A 12-lead ECG was monitored at 1 to 5 and 6 to 15 minutes after injection. The occurrence of adverse effects was sought through open-ended interview.
A 30-min washout period was then observed, followed by administration of 2% DDFP emulsion (0.05 ml/kg). Repeat echocardiographic images were recorded during the injection and until contrast effect was no longer visible in the left ventricular cavity. Blood pressure, heart rate, respiratory rate and oximetry were measured eight times during the 15 min after injection. A 12-lead ECG was monitored at 1 to 5 and 6 to 15 min after injection. The occurrence of adverse effects was sought through open-ended interview. A repeat blood sample was obtained 1 h after the injection. Each patient was asked to return 18 to 48 h after the study for a repeat blood sample, vital signs, pulse oximetry and occurrence of adverse events.
The primary efficacy evaluation was performed by two experienced echocardiographers (A.N.D., D.J.S.) who read the studies in randomized order without knowledge of whether the contrast images were obtained after DDFP or active control. The blinded readers were specifically trained in grading endocardial border definition and LV opacification using the scoring system described below. In addition, the 18 primary investigators graded each echocardiogram in an unblinded manner.
Efficacy end points
The efficacy of each contrast echocardiographic agent was assessed by means of end points that were organized into four tiers: 1) technical end points, 2) anatomic end points, 3) diagnostic end points, and 4) patient management end points.
Technical end points included the intensity and duration of contrast enhancement provided by the test agent. The intensity of LV opacification after contrast administration was graded as 0 = none; 1 = faint detectable contrast; 2 = intermediate opacification; and 3 = full opacification. The duration of LV opacification was measured (by the investigators only) from the appearance of contrast in the left ventricular until chamber opacification returned to baseline. In the subset of patients in whom the contrast agent worsened the image, the contrast imaging window was calculated as the time that contrast was present in the chamber minus the time that the image was degraded.
Anatomic end points included the ability of the test agent to improve visualization of the endocardial border and salvage a technically difficult study. Endocardial border definition was graded in each of 12 myocardial segments in the apical two- and four-chamber views for both baseline and contrast echocardiograms as 0 = border not seen; 1 = faint border delineation; 2 = intermediate border delineation; and 3 = excellent border delineation. An endocardial border delineation score was calculated as the sum of individual segment scores divided by 12 (the number of segments). Salvage of a suboptimal echocardiogram was defined as improvement from an endocardial border definition score from 0 or 1 in at least two adjacent segments to a score of 2 or 3 in at least five of six segments in both the apical two- and four-chamber views. Facilitation of wall motion assessment by the contrast agent was graded as yes or no.
Diagnostic end points were chosen to determine whether the contrast agent assisted in making a final diagnosis. The investigators and blinded readers were specifically asked whether the agent 1) provided adequate contrast enhancement of the blood pool; 2) facilitated endocardial border delineation; 3) improved the quality of regional wall motion analysis; 4) provided enhancement of the myocardium; 5) shortened the time needed to reach a diagnosis (blinded readers only); 6) improved confidence in the diagnosis and, if yes, to record the degree of improvement as minimal, low, moderate or significant.
Patient management end points evaluated the potential ability of the test agent to affect a patient’s diagnostic or therapeutic management strategy. The blinded readers and investigators were asked whether the contrast agent 1) disclosed findings that were not apparent on the baseline echocardiogram; or 2) would have eliminated the need for additional diagnostic tests or referrals.
Safety end points
Table 1shows the clinical and laboratory variables that were measured before and after DDFP. Adverse events were classified as mild, moderate or severe in intensity. The relation between adverse events and the test agent was determined by the investigator to be definitely related, probably related, possibly related, unknown or not related.
Differences between active control and DDFP in continuous variables, such as LV opacification scores and endocardial border delineation scores, were compared using paired t tests. Differences in dichotomous variables were compared by McNemar’s test. Concordance between observers was assessed by the Pearson coefficient of correlation. A p value ≤0.05 was considered statistically significant.
Two hundred fifty-four patients were enrolled (21 to 94 years old, mean age 61; 185 men [73%], 69 women [27%]; 221 whites [87%], 30 African-Americans [12%], 3 Asian-Americans [1%]). The diagnosis for which the echocardiogram was obtained was coronary artery disease in 56% of patients, myocardial infarction in 22%, congestive heart failure in 16%, valvular heart disease in 19%, arrhythmias in 12%, endocarditis in 1%, pulmonary hypertension in 1% and pericardial disease in 1%.
Figure 1 shows the percent of patients with full or intermediate opacification of the LV cavity for each contrast agent as interpreted by both the reviewers and the investigators. Full or intermediate LV cavity opacification was consistently higher for DDFP than for active control (78% vs. 31% for reader A, 69% vs. 34% for reader B, 83% vs. 55% for the primary investigators, all p < 0.0001).
Figure 2 illustrates the mean LV cavity opacification scores for both contrast agents as determined by the readers and the investigators. Scores were significantly higher for DDFP than for active control (2.3 ± 0.07 vs. 1.1 ± 0.06 for reader A, 2.0 ± 0.07 vs. 1.2 ± 0.06 for reader B, 2.4 ± 0.06 vs. 1.5 ± 0.06 for the investigators, all p < 0.001).
The duration of contrast effect was significantly longer for DDFP than for active control (4.2 ± 0.52 vs. 0.53 ± 0.02 min, p < 0.001). Contrast resulted in worsening of the LV image in 37 patient (15%) after DDFP compared with 48 patients (19%) after active control (p = NS). In these patients, the contrast imaging window was 134 s with DDFP and 8 s with active control.
Endocardial border delineation
Compared with baseline images, endocardial border delineation was improved by DDFP in 88% of patients versus 45% with active control for the investigators, 86% versus 58% for reader A and 83% versus 62% for reader B (all p < 0.001). In the four-chamber view, endocardial border detection was improved in 86% of patients with DDFP versus 47% with active control for the investigators, 83% versus 52% for reader A, and 81% versus 59% for reader B (all p < 0.001). In the two-chamber view, endocardial border detection was improved in 80% after DDFP compared with 30% with active control for the investigators, 75% versus 40% for reader A and 79% versus 50% for reader B (all p < 0.001). Concordance between observers was good for endocardial border definition (r = 0.77 between readers, r = 0.70 between readers and investigators).
The change in endocardial border delineation score from baseline to after contrast administration is shown in Figure 3. Improvement in endocardial border delineation was greater with DDFP than with active control for blinded reader A (1.1 ± 0.05 vs. 0.3 ± 0.03, p < 0.001), blinded reader B (1.2 ± 0.06 vs. 0.4 ± 0.04, p < 0.001) and the investigators (0.9 ± 0.04 vs. 0.2 ± 0.02, p < 0.001).
Salvage of suboptimal echocardiograms
Baseline echocardiograms improved from suboptimal to diagnostically useful in 51% of patients after DDFP compared with 9% after active control for the investigators, 54% versus 9% for blinded reader A and 50% versus 16% for blinded reader B (all p < 0.001). Assessment of regional wall motion was facilitated by the contrast agent in 88% of patients after DDFP compared with 37% with active control for the investigators, 85% versus 25% for blinded reader A and 92% versus 71% for blinded reader B (all p < 0.001). Figure 4 shows an example of a study that improved from suboptimal to diagnostically useful with DDFP.
Change in diagnostic confidence
The investigators reported increased confidence in the diagnosis in 76% patients after DDFP compared with 24% after active control (p < 0.001). The degree of improvement in diagnostic confidence was reported as moderate or significant in 80% of cases after DDFP compared with 43% after active control (p < 0.0001).
Patient management end points
According to the investigators, DDFP added sufficient diagnostic information to influence management end points in 162 patients (65%) with DDFP compared with 54 patients (22%) with active control (p < 0.001). Findings not apparent on the baseline study were disclosed more frequently by DDFP than by active control (63% vs. 20%, p < 0.001). Compared with active control, DDFP could have prevented the need for additional tests or referrals in 34 patients (19%) versus 12 patients (7%) with active control (p < 0.001).
Neither contrast agent resulted in significant changes in heart rate, blood pressure, respiratory rate, oxygen saturation, ECG variables or laboratory values.
A total of 89 adverse events were reported in 44 patients and were considered by the investigators to be related to DDFP in 23 patients (9%) and to active control in 16 (6%) (p = 0.92). Most adverse events were considered to be mild in severity and resolved spontaneously within 5 min (Table 2).
One patient died during the follow-up period. He was a 57-year old white man with ischemic cardiomyopathy who had been admitted to the hospital for congestive heart failure 1 week before the study. On the day of study, he was classified as having functional class II heart failure, and his serum potassium was 3.1 mEq/dl (results returned 3 days later). He had no adverse effects with either active control or DDFP and returned home. He died suddenly in his sleep ∼9 h after the study. The investigators ruled his death to be unrelated to the test agents and considered it to be an arrhythmic death secondary to underlying heart disease and hypokalemia.
Evaluation of regional and global LV function is one of the major clinical indications for echocardiography (13). However, dropout of the endocardial border in apical views is common and may preclude assessment of LV function even in patients with adequate acoustic windows (14). In a recent study of post–myocardial infarction patients, full visualization of all endocardial borders in the apical four- and two-chamber views could only be accomplished in 18 (36%) of 50 patients (9). Therefore, the ability to adequately define the endocardial borders by using a transpulmonary contrast agent is an important clinical goal in echocardiography (5–10). Although the active control used in the present study is the only FDA-approved agent currently available for this indication, its use has been limited by several factors, including microbubble destruction and poor persistence in the circulation (15–18). Our study demonstrates that 2% DDFP emulsion (EchoGen) is safe and is superior to sonicated 5% human albumin (Albunex) in intensity of LV opacification, duration of LV opacification and endocardial border definition. Consequently, the investigators and blinded readers considered that this new agent could increase confidence in the diagnosis and potentially improve management strategy.
In a previous large clinical trial of sonicated 5% human albumin (9), LV opacification was judged to be full or intermediate in 74% by independent reviewers. In the present study, the same agent resulted in intermediate or full LV opacification in 55% of patients. The reasons for these different results are unclear; however, several possibilities exist: 1) The patients in our study tended to be older than those in the previous trial (mean age 61 vs. 56 years). 2) There may be unmeasured differences between the patient groups in clinical variables or echocardiographic technique. For example, low cardiac output has been associated with increased microbubble destruction (18) and was not specifically reported in either study. Neither study specified acoustic power or other instrument settings but merely required that they be optimized for each patient. It is now known that increased acoustic power and continuous ultrasound delivery enhance destruction of encapsulated microbubbles (18,19). Despite these considerations, the study design, wherein each patient was used as his own control and the data were interpreted in blinded manner by independent readers, justifies the conclusion that DDFP is superior to sonicated 5% human albumin for LV opacification in humans.
Even in cases where sonicated 5% human albumin fully opacified the LV cavity, the duration of effect was generally too short to allow the sonographer to obtain complete images of the LV from multiple acoustic windows. In contrast, the prolonged duration of contrast effect with DDFP facilitated the use of multiple views after a single injection. This prolonged duration of effect is largely due to the low diffusivity and blood solubility of perfluorocarbons relative to air. In addition, it is conceivable that encapsulated microbubbles, such as sonicated 5% human albumin, are more prone to destruction than nonencapsulated agents, such as 2% DDFP emulsion. It has been noted that air-filled microbubbles demonstrate a decrease in contrast intensity in the LV cavity during systole; this has been attributed to microbubble destruction by intracavitary pressure (17,20) and acoustic pressure from the ultrasound beam (18). In the present study, these effects were noted to be much less pronounced with DDFP, suggesting that this agent may be less subject to microbubble destruction.
Another factor contributing to the longer contrast imaging window of DDFP was that it caused less attenuation of far-field structures than did sonicated albumin. The mechanism of this beneficial effect is thought to be related to lower diffusion rates of higher molecular weight gases out of the microbubble as well as their low blood solubility (21). These effects allow improved contrast effect at lower doses and thereby reduce attenuation artifacts (21).
Endocardial border definition
Accurate assessment of LV volumes, ejection fraction and regional wall motion is dependent on adequate visualization of the endocardial surface (13,14). The present study shows that compared with sonicated 5% human albumin, DDFP provides significantly greater improvement in endocardial border delineation in patients with a suboptimal echocardiogram. Although it is reasonable to assume that this would translate into improved accuracy in quantitative assessment of LV global and regional wall motion, we did not compare the contrast-enhanced echocardiograms with an independent reference standard. However, Hundley et al. (6) recently showed that 2% DDFP emulsion improved the accuracy of echocardiographically measured LV volumes, ejection fraction and regional wall motion scoring compared with cine magnetic resonance imaging.
Studies in open chest dogs (22,23) have shown that myocardial perfusion defects in the setting of acute myocardial ischemia can be detected by intravenous injection of DDFP. In those studies, contrast effect persisted in the myocardium even after it cleared the LV cavity, suggesting that some of the microbubbles remained in the myocardium. In the present study, myocardial contrast enhancement was not consistently observed with either agent. The reasons for the apparent differences in myocardial contrast enhancement between the canine and human studies are not known but are probably related to the obvious differences between anesthetized, open chest, mechanically ventilated dogs without intrinsic myocardial or coronary artery disease and conscious patients with a variety of cardiovascular or systemic diseases. Dogs also received a higher dosage of DDFP. In addition, we did not use second harmonic or intermittent imaging, modalities that have been shown to greatly augment myocardial enhancement in humans (19,24–26).
A previous canine study (22) demonstrated pulmonary hypertension and hypoxemia at high doses (0.7 ml/kg) of DDFP. Subsequently, a change in the formulation of the emulsion and the use of hypobaric activation procedures allowed studies to be done at much lower doses than that used in our study (0.05 ml/kg). At this dose we found that DDFP was safe and well tolerated. The side effect profile was similar to that of sonicated albumin, and there were no significant changes in vital signs, oximetry, ECG findings or laboratory values.
The present study was performed using fundamental imaging at frame rates of at least 30/s. The development of second harmonic imaging (18,19,24–26) and ECG triggering at 1 frame/cycle (19) improved the visualization of contrast in the LV cavity and myocardium. It is likely that the scores for both agents would have been higher had these newer imaging modalities been used. Nevertheless, because most clinical echocardiography laboratories do not have second harmonic capability at this time, the results of the present study are applicable to the current practice of echocardiography.
This Phase III multicenter trial demonstrates that in patients with a suboptimal echocardiogram, DDFP emulsion (EchoGen) is superior to sonicated human albumin (Albunex) for LV cavity opacification, endocardial border definition, duration of effect, salvage of suboptimal echocardiograms, diagnostic confidence and potential to affect patient management. The two agents had similar safety profiles. These findings may not be applicable to patients with clinically active pulmonary disease or obesity-related sleep apnea.
Participating institutions and investigators for the EchoGen Contrast Echo Study Group
Clinical centers.University of Texas Southwestern Medical Center, Dallas, Texas: Paul A. Grayburn, MD, Michael L. Main, MD; Johns Hopkins University School of Medicine, Baltimore, Maryland: James L. Weiss, MD, Elizabeth Klodas, MD; Deaconess-Nashoba Hospital, Ayer, Massachusetts: Terrence C. Hack, MD; Thomas Jefferson University, Philadelphia, Pennsylvania: Joel S. Raichlen, MD; New England Medical Center, Boston, Massachusetts: Manni A. Vannan, MD; Cleveland Clinic Foundation, Cleveland, Ohio: Allan L. Klein, MD; Bowman Gray University Medical Center, Winston-Salem, North Carolina: Dalane W. Kitzman, MD; Oklahoma Cardiovascular and Hypertension Center, Oklahoma City, Oklahoma: Steven G. Chrysant, MD, FACC; Veterans Affair’s Medical Center, East Orange, New Jersey: Jerald L. Cohen, MD; University of California at Irvine, Orange, California: David Abrahamson, MD; University of California at San Francisco, San Francisco, California: Elyse Foster, MD; Washington University Medical School, Saint Louis, Missouri: Julio E. Perez, MD; University of Massachusetts, Worchester, Massachusetts: Gerard P. Aurigemma, MD; National Heart, Lung, and Blood Institute, Bethesda, Maryland: Julio A. Panza, MD; Massachusetts General Hospital, Boston, Massachusetts: Michael H. Picard, MD; Vanderbilt University, Nashville, Tennessee: Benjamin F. Byrd III, MD; Indiana University, Indianapolis, Indiana: Douglas S. Segar, MD; Research for Health, Inc., Houston, Texas: Stuart A. Jacobson, MD.
Core echocardiography laboratories. University of Oregon Health Science Center, Portland, Oregon: David J. Sahn, MD; University of California, San Diego, San Diego, California: Anthony N. DeMaria, MD.
☆ This study was funded by SONUS Pharmaceuticals, Bothell, Washington.
- electrocardiogram, electrocardiographic
- Food and Drug Administration
- left ventricular
- Received November 10, 1997.
- Revision received March 12, 1998.
- Accepted April 8, 1998.
- American College of Cardiology
- Kisslo J
- Krumholz H.M,
- Douglas P.S,
- Goldman L,
- Waksmonski C
- Hundley W.G,
- Kizilbash A,
- Afridi I,
- Franco F,
- Peshock R.M,
- Grayburn P.A
- Feinstein S.B,
- Cheirif J,
- TenCate F.J,
- et al.
- Crouse L.J,
- Cheirif J,
- Hanly D.E,
- et al.
- Porter T.R,
- Xie F
- Cotter B,
- Kwan O,
- Kimura B
- Cheitlin M.D,
- Alpert J.S,
- Armstrong W.F,
- et al.
- Wei K,
- Skyba D.M,
- Firschke C,
- Lindner J.R,
- Jayaweera A.R,
- Kaul S
- Porter T.R,
- Xie F
- Shapiro J.R,
- Reisner S.A,
- Lichtenberg G.S,
- Meltzer R.S
- Porter T.R,
- Xie F,
- Kricsfeld A,
- Deligonal U,
- Kilzer K
- Grayburn P.A,
- Erickson J.M,
- Escobar J.F,
- Womack L,
- Velasco C.E
- Porter T.R,
- Li S,
- Kricsfeld D,
- Armbruster R.W
- Kaul S,
- Senior R,
- Dittrich H,
- Raval U,
- Khattar R,
- Lahiri A