Author + information
- Received May 29, 1998
- Revision received May 27, 1999
- Accepted August 30, 1999
- Published online December 1, 1999.
- Andrew D. Maslow, MD∗∥,* (, )
- Meredith M. Regan, ScD†,
- J.Michael Haering, MD∗,
- Robert G. Johnson, MD‡ and
- Robert A. Levine, MD§
- ↵*Reprint requests and correspondence: Dr. Andrew D. Maslow, Department of Anesthesia, Rhode Island Hospital, 593 Eddy Street, Davol 129, Providence, Rhode Island 02903.
To determine predictors of systolic anterior motion and left ventricular outflow tract obstruction (SAM/LVOTO) after mitral valve repair (MVRep) in patients with myxomatous mitral valve disease.
Mechanisms for the development of SAM/LVOTO after MVRep have been described; however, predictors of this complication have not been explored. We hypothesize that pre-MVRep transesophageal echocardiography (TEE) can predict postrepair SAM/LVOTO.
Using TEE, the lengths of the coapted anterior (AL) and posterior (PL) leaflets and the distance from the coaptation point to the septum (C-Sept) were measured before and after MVRep in 33 patients, including 11 who developed SAM/LVOTO (Group 1) and 22 who did not (Group 2).
Group 1 patients had smaller AL/PL ratios (0.99 vs. 1.95, p < 0.0001) and C-Sept distances (2.53 vs. 3.01 cm, p = 0.012) prior to MVRep than those in Group 2. Resolution of SAM/LVOTO was associated with increases in AL/PL ratio and C-Sept distance. This reflects a more anterior position of the coaptation point in those who developed SAM/LVOTO.
These data suggest that TEE analysis of the mitral apparatus can identify patients likely to develop SAM/LVOTO after MVRep for myxomatous valve disease. The findings are consistent with the concept that SAM of mitral leaflets is due to anterior malposition of slack mitral leaflet portions into the LVOT. The position of the coaptation point of the mitral leaflets is dynamic and a potential target and end point for surgical designs to prevent SAM/LVOTO post MVRep.
Systolic anterior motion of the mitral valve and left ventricular outflow tract obstruction has been described after mitral valve repair (MVRep) in patients with myxomatous mitral valve disease (1–8). The incidence ranges from 2% to 16%, despite recent advances in MVRep (1,5,9).
It has been suggested that systolic anterior motion of the mitral leaflets (SAM) and left ventricular outflow tract obstruction (LVOTO) may not be entirely or primarily due to the Venturi effect (10–12). Studies have related SAM to abnormalities of the mitral valve apparatus itself (5,10,12–15). Contributing factors include the position of papillary muscles and mitral leaflet anatomy (10,12,13,15). Anteriorly displaced papillary muscles malposition the mitral valve toward the left ventricular outflow tract, increasing the likelihood of SAM and LVOTO (5). This increased risk may be due to changes in direction of ventricular flows and the presence of mitral tissue in the outflow tract, making it more susceptible to systolic outflow (10,12,13,15). Other causes include abnormalities of the mitral leaflets (5,14). In particular, a relatively large posterior leaflet may coapt with the anterior leaflet closer to its base and cause both an anterior shift of the coaptation point, and an increase in the amount of slack leaflet tissue in the outflow tract (5,14). The residual leaflet portion beyond the coaptation point, unlike the coapted portions that are held in place by the transmitral pressure difference, are relatively free to move in response to flow-related forces in the LVOT (16). Elongation of the anterior leaflet may cause a similar increase in the relatively slack residual leaflet. Both of these conditions make the leaflets more susceptible to the effects of systolic outflow (14). In vitro analysis has shown that increases in both anterior and posterior leaflet positions can predispose to systolic anterior motion (14). Proposed mechanisms for SAM postmitral valve repair have therefore incorporated the concept that the leaflet anatomy and position contribute to systolic anterior motion (5,6).
The purpose of this study was to explore prerepair echocardiographic predictors of SAM and LVOTO after MVRep. We hypothesized that prerepair transesophageal echocardiographic (TEE) evaluation has the potential to predict the development of systolic anterior motion of the mitral leaflets and LVOTO after cardiopulmonary bypass in patients undergoing MVRep for myxomatous valve disease based on features of the mitral apparatus that would predispose to SAM. This predictive information could potentially allow tailoring of the surgical approach to valve repair designed to limit or avoid this complication.
We studied a total of 33 patients who were undergoing TEE evaluation before MVRep for myxomatous mitral valve disease with prolapse, as defined by Perloff et al. (17). According to the database, the overall incidence of SAM/LVOTO after repair in this patient population was approximately 5%. These included 11 consecutive patients who developed SAM and LVOTO after MVRep (Group 1) and 22 consecutive patients in whom this did not occur (Group 2). These were not 33 consecutive MVRep cases.
Table 1shows demographic data including age, gender, surgical procedure, valve morphology and pre- and postcardiopulmonary fractional shortening. Fractional shortening was measured using the following formula:
Image acquisition and measurement
Intraoperative TEE studies were performed in standard views, particularly longitudinal views of the left heart from the left atrial window, using a Hewlett-Packard phased array sector scanner. Studies were reviewed and measured by two examiners (A.D.M. and J.M.H.). One examiner was aware of the presence of SAM after repair (A.D.M.) and the second (J.M.H.) was not. Clinically and hemodynamically important systolic anterior motion, as defined by Grigg et al. using two-dimensional and color Doppler, was said to be present if there was:
1. two-dimensional visualization of systolic anterior motion into the left ventricular outflow tract, and
2. a mosaic pattern by color Doppler seen in the left ventricular outflow tract, suggesting increased velocities and disturbed or turbulent flow and in the left atrium, consistent with mitral regurgitation (18). Measurements were made pre- and postcardiopulmonary bypass in Groups 1 and 2, as well as after resolution of SAM/LVOTO in Group 1. Each examiner made each measurement three times from three cardiac cycles. Measurements from both examiners were averaged.
Measurements were made from the transverse five-chamber view, as reported by Lee et al. (5). Visualization of mitral valve coaptation and the left ventricular outflow tract can be consistently obtained in this view (Fig. 1). In order to evaluate the geometry of the mitral valve and its relation to the left ventricular outflow tract and ventricular cavity, all measurements were made at the onset of ventricular ejection. As shown in Figure 1, we measured the lengths of the coapted anterior (AL) and posterior (PL) leaflets (annulus to coaptation for each leaflet) to explore the contribution of each leaflet to coaptation, the minimum distance from the coaptation point to the septum (C-Sept), the diameter of the mitral valve annulus (Ann Diam) and the length of the residual anterior leaflet portion beyond the coaptation point. All diameter measurements were made using leading edge to leading edge. After placement of the annulus, the annular diameter was made from the inner aspects of the ring. The LVID was measured at the same time in systole in the same view (LVIDs). As shown in Figure 1, the diameter of the LV was measured at the base of the heart instead of the midventricular level. This was done for two reasons. The first was because this was the more consistently obtained site. The second, and more important reason, was to measure the dimension at the site most likely to contribute to SAM/LVOTO. The perpendicular axial distance from the coaptation point to the mitral annulus (Coapt-Ann) was also measured to express how far leaflet coaptation is positioned into the left ventricle.
During SAM the measured anterior leaflet length was the straight distance from the anterior annulus to the site of coaptation. The C-Sept distance was measured as the minimum distance from the coaptation to the septum.
In cases of flail leaflet a series of cardiac cycles was obtained for each measurement. Scrolling through the cycle allowed the examiners to estimate where the leaflets may have coapted during systole. This spot was taken as the site of coaptation and subsequent measurements were made using this reference.
Mean, standard deviation and range of data points are presented in Table 2. Prerepair measurements were compared between Group 1 and Group 2 patients using Student two-sample ttests; the corresponding p values are also reported. Because we are examining multiple end points, we consider p < 0.01 as statistically significant.
If for any measurement there was no overlap of the ranges of values of the two groups so that whether or not the subject developed SAM/LVOTO was perfectly predicted by the measurements, we calculated 95% tolerance intervals (or prediction intervals for a future observation). While the range represents the limits of the observed sample values, the 95% tolerance interval represents the expected limits of 95% of the corresponding population values; that is, the interval is expected to enclose at least 95% of the population values.
As a secondary analysis, the groups were compared with respect to their measurements after repair to examine whether differences between the groups persisted and how the measurements changed from pre- to postrepair. This was done to identify possible mechanisms of SAM/LVOTO, as was done in a previous study (5). Measurements after repair were compared between groups using two-sample ttests. The corresponding p values are also reported in Table 2. Changes between pre- and postrepair were compared using paired ttest, separately, for each group for simplicity; the general significance levels were denoted in Table 2using symbols. Because of the multiplicity of testing for each measurement, we considered p < 0.01 as statistically significant.
Interobserver variability was assessed for prerepair measures using bias analysis. The differences between measurements made by two independent examiners were divided by the means and multiplied by 100%. The results were averaged to give a mean bias reported as a percent difference.
A total of 33 patients were studied in the two groups; they did not have significant differences in age, gender, pre- and postrepair fractional shortening, anterior versus posterior leaflet involvement in prolapse, surgical procedure or use of vasoactive drugs to facilitate separation from cardiopulmonary bypass (Table 1). All patients in both groups underwent mitral valve ring annuloplasty and quandrangular resection of the posterior leaflet with no surgical procedure performed on the anterior leaflet.
Precardiopulmonary bypass data (Table 2)
Precardiopulmonary bypass data showed a significant difference between Groups 1 and 2 with respect to the contribution of the valve leaflets to coaptation. The position of the coaptation point (C-Sept distance) tended to lie closer to the LVOT in Group 1 (p = 0.012). Group 1 patients had a smaller AL/PL ratio (0.99 vs. 1.95; p = 0.0001) and C-Sept distance (2.53 cm vs. 3.01 cm; p = 0.012) compared with Group 2. Group 1 patients had a larger PL (2.18 cm vs. 1.42 cm; p < 0.0001) and smaller AL (2.13 cm vs. 2.68 cm; p = 0.0004). There was no overlap in the prerepair ratio of coapted anterior to posterior leaflet length between those who developed SAM/LVOTO (0.75–1.17) and those who did not (1.33–2.80), reflecting a more anterior position of coaptation in those who developed SAM. The corresponding 95% tolerance intervals indicate that we would expect at least 95% of patients who develop SAM/LVOTO to have AL/PL ratios between 0.66 and 1.32 and at least 95% of patients who do not to have AL/PL ratios between 1.01 and 2.89, suggesting that there is limited overlap in the populations also. Figures 2 and 3⇓⇓demonstrate images of patients from Groups 1 and 2.
Changes in echocardiographic measurements from before to immediately after cardiopulmonary bypass (Table 2)
Both groups showed decreases in the lengths of the coapted anterior and posterior leaflet portions, annular diameters and C-Sept distance from before to after bypass with increases in the amount of residual leaflet measured beyond the coaptation point. Group 2 patients showed a decrease in LVIDs, while Group 1 patients showed little change. The AL/PL ratio in Group 2 patients increased by nearly 60% (relative posterior shift in the coaptation point), while this measurement stayed essentially unchanged in Group 1.
Postcardiopulmonary bypass measurements (Table 2)
Postcardiopulmonary bypass measurements showed that differences between Group 1 and Group 2 persisted. Group 1 continued to have smaller AL/PL ratios (1.14 vs. 3.08), smaller C-Sept distances (1.90 cm vs. 2.52 cm) and greater residual leaflet (1.61 cm vs. 0.68 cm) than Group 2. Group 1 patients also had a greater distance from the coaptation point to the annulus in the immediate postbypass period (0.84 cm vs. 0.47 cm) with a tendency for an increase in this distance immediately after repair compared with a decrease in Group 2.
Of the 11 patients in whom SAM and LVOTO was seen, four had complete resolution of LVOTO and mitral regurgitation after weaning inotropes, fluid administration and, in one patient, administration of esmolol. With resolution of SAM, these four patients showed increases in AL/PL ratio and C-Sept distance with a decrease in the length of the residual tissue beyond the coaptation point (Table 3). The coaptation point to the mitral annulus (Coapt-Ann) also decreased with resolution of SAM.
Of the 11 patients with SAM, 7 still had SAM after pharmacologic and fluid therapy, four without color Doppler evidence of LVOTO and mitral regurgitation. The three patients with clinically significant SAM were compared with the eight without residual LVOTO or moderate mitral regurgitation (Table 4). Although all of these patients had a relatively anterior position of coaptation (low AL/PL) prerepair, those with persistent outflow tract obstruction had larger anterior and posterior leaflet lengths (AL: 2.46 vs. 2.01 cm, PL: 2.45 vs. 2.08 cm), providing increased leaflet tissue available for SAM.
Interobserver variability was 4.4% on average for all prerepair measurements (Table 5)with a mean difference of 0.03 cm.
We have demonstrated that echocardiographic evaluation of the mitral valve may predict development of SAM/LVOTO in patients with myxomatous disease undergoing MVRep. Relatively greater contribution of the posterior leaflet (lower AL/PL) to the coaptation of the mitral valve prior to repair was uniformly found in patients who exhibited SAM/LVOTO after repair. Also, those patients with SAM/LVOTO (Group 1) tended to have a smaller prerepair C-Sept compared with those without (Group 2). These differences persisted after repair during SAM/LVOTO.
In four Group 1 patients, resolution of postrepair systolic anterior motion was associated with a decrease in PL, increased AL/PL, increased C-Sept, reduction in residual leaflet and movement of the coaptation point posteriorly and closer to the annular plane.
Three Group 1 patients with longer anterior mitral leaflets prior to repair and a greater reduction in annular diameter after repair were likely to show color Doppler evidence of outflow tract obstruction despite treatment with intravenous fluid, vasopressors and discontinuation of inotropes.
These data demonstrate that the posterior and anterior leaflet contribution to valve closure and position of the coaptation point contribute to SAM of the mitral leaflets. The larger posterior leaflet coapts with the anterior leaflet closer to its base resulting in increased residual or “slack” leaflet portions which lie closer to the outflow tract and are thus more susceptible to systolic outflow.
In Group 1 the postrepair AL/PL ratio was less than 1.2 despite a posterior leaflet plication. Anterior leaflet length decreased, while leaflet beyond the coaptation point increased. The reduction in AL is probably due to coaptation toward its base with a relatively large PL and reduction in annular diameter. In addition, the distance from the coaptation point to the mitral annular plane (C-Ann) in Group 1 increased compared with Group 2. This resulted in increased slack leaflet and coaptation close to the left ventricular outflow tract and further into the left ventricular cavity (shorter C-Sept and larger C-Ann). Such changes in coaptation increase the likelihood of SAM/LVOTO.
Relation to literature
This is the first study to investigate echocardiographic predictors of SAM and LVOTO in this population. There have been studies examining the mechanism of SAM after MVRep. Lee et al. also demonstrated an increase in posterior leaflet length (posterior annulus to coaptation point) and a smaller C-Sept distance in the period after bypass in patients with SAM and LVOTO (5). C-Sept distances prior to repair and during SAM were similar in the two studies (2.6 cm and 1.7 cm, respectively, in Lee et al. and 2.5 cm and 1.9 cm in our study). Also, prerepair posterior leaflet length in Lee’s study was 1.9 cm compared with 2.2 cm in our study. As in our study, resolution was associated with decreased posterior leaflet contribution to coaptation and an increased C-Sept distance (2.3 cm in Lee et al. vs. 2.4 cm above). These changes occurred after partial or complete removal of the annuloplasty resulting in complete resolution in 7 of 14 patients and significant improvement in the other 7. The authors concluded that “an annuloplasty of any kind can displace the valve coaptation into the left ventricle.”
In several experimental models, alterations in the mitral apparatus caused SAM and LVOTO (10,12–15). Cape et al. used an in vitro flow chamber to demonstrate that anterior and inward displacement of the papillary muscle altered chordal tension resulting in residual leaflets and systolic anterior motion (10). Levine et al. demonstrated that anterior displacement of the papillary muscles shifted the mitral coaptation point toward the base of the leaflets, resulting in systolic anterior motion (12). Lefebvre et al. demonstrated that anteriorly placed papillary muscles resulted in systolic anterior motion in the absence of hypertrophy (15). However, with a posterior placed papillary muscle, no systolic anterior motion was seen despite severe septal hypertrophy and flow velocities of 3.3 m/s. Using an in vitro pulsatile flow model, Lefebvre et al. showed that displacement of papillary muscle affected ventricular flow direction (13). With normal papillary muscle position, systolic outflow was proximate to the septum and exerted minimal forces on the mitral leaflets. However, anterior placement of the papillary muscle with subsequent movement of the mitral apparatus toward the left ventricular outflow tract altered ventricular flows such that systolic flows were proximate to the posterior wall, impacting on the mitral leaflets and resulting in systolic anterior motion.
He et al. (14), using a similar in vitro pulsatile flow model, demonstrated that leaflet elongation favored SAM by creating long overlapping residual leaflets capable of moving anteriorly. In addition, posterior leaflet elongation favored SAM by shifting leaflet coaptation anteriorly. Residual leaflet length correlated with the degree of SAM.
With the findings of this study and those of Lee et al., it should be expected that the “sliding leaflet technique,” in addition to a quadrangular resection, proposed by Jebara et al., should reduce the incidence of SAM and LVOTO (1,5). By decreasing the size of the posterior mitral leaflet, coaptation will occur closer to the tip of the anterior leaflet and more posteriorly in the ventricular cavity, leaving less mitral tissue near the left ventricular outflow tract.
Reed et al. (19)presented two patients with SAM/LVOTO after mitral valve repair who were treated with transaortic resection of redundant anterior leaflet. Anterior leaflet plication has subsequently been suggested by Grossi et al. who demonstrated a reduction in SAM/LVOTO when used in addition to posterior leaflet surgery (20,21).
A closer inspection of data presented by Jiang et al. (11)demonstrates that lengths of the coapted anterior and posterior leaflets and the AL/PL ratio in patients with hypertrophic obstructive cardiomyopathy (HOCM) and systolic anterior motion were similar to our Group 1 patients before and during SAM/LVOTO (HOCM; AL 1.8 cm, PL 1.9 cm, AL/PL 1.0 vs. Group 1 [prerepair]; AL 2.1, PL 2.2, AL/PL 1.0 and Group 1 [during SAM] AL 1.4 cm, PL 1.2 cm, AL/PL 1.1). Measurements of AL/PL in both normal patients and patients with nonobstructive hypertrophic cardiomyopathy (HCM) in Jiang et al. (11)were similar to Group 2 patients (2.0, 2.3 and 2.0, respectively). These comparisons suggest that defined and measurably similar leaflet contributions to coaptation are associated with SAM across different populations.
In other studies of patients with HOCM, AL/PL ratios were similar to our Group 2 patients (22,23). These results suggest that a long or redundant anterior leaflet increased the risk of systolic anterior motion. Methodological differences may account for these findings, as mitral leaflet measurements were made during diastole or from pathology specimens with the leaflets spread out to maximum dimensions. In contrast, we measured mitral dimensions during systole to assess the functional or physiological dimensions during this critical phase in the cardiac cycle. These studies do, however, suggest that a large anterior leaflet contributes to SAM. This was also suggested in Group 1 patients in which SAM/LVOTO persisted despite medical treatment.
Kofflard et al. (24)performed myomectomy with anterior leaflet extension using a gluteraldehyde preserved autologous pericardial patch. The patch was used to “stiffen” the anterior leaflet thus reducing leaflet laxity (24,25). These patients were favorably compared with 12 patients undergoing myomectomy alone. Eliminating this laxity of the anterior leaflet may decrease the effects of systolic outflow on the leaflets and reduce anterior motion and outflow tract obstruction. Anterior leaflet plication, with or without myomectomy, has also been reported by McIntosh et al. (26)in patients with HOCM. These studies support the idea that abnormalities of the mitral valve contribute significantly to outflow tract obstruction and may be the ideal target for surgical treatment of HOCM (24,26).
The results presented in this article, along with a growing body of literature, emphasize the importance of the mitral valve apparatus as a cause of SAM/LVOTO. Simple precardiopulmonary bypass mitral valve measurements, as presented above, may help guide surgical repair and prevent SAM/LVOTO. While the length of the PL and position of coaptation are important targets, the length of the anterior leaflet, the AL/PL ratio and perhaps the amount of tissue beyond the coaptation point are important as well.
Data gathered retrospectively may be flawed. However, given the consistency with other studies and the growing literature relating the mitral apparatus to SAM/LVOTO, we feel that our results will be confirmed in future prospective studies (5,14).
In review of the data, there is a question of whether the Group 1 and Group 2 patients had similar procedures. Given that the repairs were done by different surgeons, it is likely that there were variations. In Group 1, there was approximately a 44% reduction in posterior leaflet height or 0.95 cm. In Group 2 there was a 57% reduction in posterior leaflet height or 0.81 cm. Although the percent reduction was greater in Group 2, the amount of posterior leaflet resected appears to have been greater in Group 1. It is possible to argue that Group 2 patients received a “better” surgery and therefore were less likely to develop SAM/LVOTO after repair; however, more posterior leaflet was, on average, resected in Group 1. Although there is probably some variation on the surgical repair, we feel that the prerepair data can identify a group of patients at risk for SAM/LVOTO and should be used to guide the surgical procedure.
Three-dimensional imaging may help further clarify the anatomy and role of the mitral valve in SAM/LVOTO. While three-dimensional analysis is well described in the literature, it is unavailable in most clinical settings. In contrast, the echocardiographic image described in this article is easily obtained.
The data collection did not include the intraoperative anesthetic record, i.e., our data did not include pulmonary artery pressures nor pulmonary capillary wedge pressures. This may have been helpful in the assessment of ventricular filling. It was our feeling that the measurement of LVIDs at the base of the ventricle would be the best assessment of cavitary dimensions that would contribute to SAM/LVOTO. Furthermore, there was no significant difference between the LVIDs between Groups 1 and 2. In addition, the decrease in LVIDs was greater for Group 2 than Group 2, making Group 2 more likely to have had SAM. However, this was not the case.
These results suggest that TEE analysis of the mitral apparatus can identify patients likely to develop SAM/LVOTO after mitral valve repair for myxomatous valve disease. The AL/PL ratio was significantly lower among patients who developed SAM/LVOTO than among those who did not. Further, the ranges of AL/PL ratios observed in the two groups of patients did not overlap, and there was minimal overlap of the corresponding 95% tolerance intervals, suggesting that the measurement may differentiate these two patient groups. Resolution was associated with shift of the coaptation point away from the left ventricular outflow tract and a greater contribution of the anterior leaflet to valve closure. Thus, our data suggest the need for a greater reduction in the posterior leaflet height if the prerepair AL/PL ratio is 1.3 or lower. The importance of this value may be greater if the C-Sept distance is 2.5 cm or less. These findings are consistent with the concept that SAM is due to anterior malposition of slack mitral leaflet portions into the left ventricular outflow tract. The position of the coaptation point of the mitral leaflets is dynamic and offers a target for surgical techniques to prevent SAM with LVOTO after mitral valve repair.
☆ This study was supported in part by grant HL38176 of the NIH, Bethesda, Maryland.
- anterior leaflet length to mitral valve contribution (annulus to coaptation)
- Ann Diam
- mitral valve annulus diameter
- distance from the coaptation point to the mitral annular plane
- coaptation point to the mitral annulus
- distance from septum to mitral valve coaptation point
- hypertrophic cardiomyopathy
- hypertrophic obstructive cardiomyopathy
- left ventricular internal diameter in systole (diastole)
- left ventricular outflow tract obstruction
- mitral valve repair
- posterior leaflet length to mitral valve contribution (annulus to coaptation)
- systolic anterior motion of the mitral leaflet(s)
- transesophageal echocardiography, echocardiographic
- Received May 29, 1998.
- Revision received May 27, 1999.
- Accepted August 30, 1999.
- American College of Cardiology
- Jebara V.A.,
- Mihaileanu S.,
- Acar C.,
- et al.
- Kronzon I.,
- Cohen M.,
- Winer H.,
- et al.
- Freeman W.K.,
- Schaff H.V.,
- Khandheria B.K.,
- et al.
- Lee K.S.,
- Stewart W.J.,
- Lever H.M.,
- et al.
- Rey M.J.,
- Mercier L.A.,
- Castonguay Y.
- Cape E.G.,
- Simon D.,
- Jimoh A.,
- et al.
- Levine R.A.,
- Vlahakes G.J.,
- Lefebvre X.,
- et al.
- Lefebvre X.P.,
- Yoganathan A.P.,
- Levine R.A.
- Grigg L.E.,
- Wigle E.D.,
- Williams W.G.,
- et al.
- Grossi E.A.,
- Galloway A.C.,
- Delianides J.,
- et al.
- Klues H.G.,
- Proschan M.A.,
- Dollar A.L.,
- et al.
- Klues H.G.,
- Roberts W.C.,
- Maron B.J.
- Kofflard M.J.,
- Van Herwerden L.A.,
- Waldstein D.J.,
- et al.
- Schwammenthal E.,
- Levine R.A.
- McIntosh C.L.,
- Maron B.J.,
- Cannon R.O. III.,
- Klues H.G.