quantitative two-dimensional echocardiographic assessment of patients undergoing left ventricular...

Quantitative two-dimensional echocardiographic assessment of patients undergoing left ventricular aneurysmectomy

To evaluate the role of quantitative two-dimensional echocardiography (2DE) in the preoperative assessment of patients undergoing left ventricular (LV) aneurysmectomy, we identified 37 patients who were studied with 2DE 1 to 56 (mean 12.6) days prior to surgery. Diastolic (Dd) and systolic (Ds) minor-axis dimensions at the base were measured and fractional shortening (FS) was calculated. Global and basilar half ejection fraction (EF) as measured from right anterior oblique left ventriculograms. At follow-up (mean 17.9 months), 27 patients were alive and clinically improved (group A) and 10 patients either died or were symptomatically unimproved (group B). Basilar half EF was significantly greater among patients in group A (0.50 + 0.09) than in group B (0.37 f 0.10) (p < 0.001). Echocardiographic FS provided the best separation between groups. Mean FS was 0.25 +- 0.06 in group A and 0.15 k 0.04 in group B (p < 0.001). All seven patients with FS < 0.17 were in group B while 25 of 27 patients with FS > 0.17 were in group A (p < 0.001). Considering all patients, basilar half EF and FS were highly correlated (r = 0.84). (AM HEART J 111:714, 1986.)

Thomas Ryan, M.D., Olivera Petrovic, M.D., William F. Armstrong, M.D., James C. Dillon, M.D., and Harvey Feigenbaum, M.D. Indianapolis, Ind.

Development of a left ventricular (LV) aneurysm is a relatively common complication of acute myocardial infarction. Surgical resection can result in improved hemodynamics’ and relief of symptoms2 but carries substantial operative risk; late mortality rates as high as 50% have been reported.3-6 Several investigators, 3*7-g by means of angiographic techniques, have demonstrated that surgical outcome can be predicted by preoperative assessment of the level of contractility of the residual myocardium. Previous work from our laboratory,‘O with the use of M-mode echocardiography, has shown that diastolic minor-axis dimensions at the mid-LV level are

From the Department of Medicine, Indiana University School of Medicine and the Krannert Institute of Cardiology.

Supported in part by the Herman C. Krannert Fund, Indianapolis, Ind.; grants HL-06308 and HL-071820, and Clinical Investigator Award HL- 01041-02 from the National Heart, Lung, and Blood Institute, national Institutes of Health, Bethesda, Md.; the American Heart Association, Indiana Affiliate; a grant from the Whitaker Foundation, Camp Hill, Pa.; a Fulbright Scholarship from the Council for International Exchange of Scholars, Washington, D. C. (Dr. Petrovic); and a Clinical Investigator Award from the National Institutes of Health/National Heart, Lung, and Blood Institute (Dr. Armstrong).

Received for publication Aug. 16, 1985; accepted Sept. 20, 1985.

Reprint requests: Thomas Ryan, M.D., Indiana University Medical Center llH N-563, 926 West Michigan St., Indianapolis, IN. 46223.

greater among patients who do not survive surgical resection. Although these M-mode echocardiographic dimensions may be helpful in predicting immediate survival following aneurysmectomy, certain lim- itations restrict the clinical utility of this technique. First, the application of M-mode echocardiography to these patients has largely been supplanted by two-dimensional techniques, which provide greater sensitivity for the detection of the aneurysm” and a means of determining the feasibility of surgical resection.12 Second, M-mode dimensions, per se, provide no direct information on residual myocardial contractility. Finally, M-mode echocardiographic measurements frequently overestimate minor-axis dimensions.13 Two-dimensional echocardiography (2DE), by providing spatial orientation, offers a more accurate and reproducible means of obtaining true dimensions.13

The purpose of the current study was to determine if routine measurements derived from 2DE could be used to predict outcome following LV aneurysmectomy. Based on the results of previous work which related outcome to residual myocardial function, this investigation was undertaken to explore three questions. First, is residual myocardial function more important than global LV function in

714

Volume 111

Number 4 2DE prior to aneurysmectomy 715

Fig. 1. Group A. Parasternal long-axis views at end-diastole (DIAST) and end-systole (SYST) are shown from a patient with an aneurysm involving the distal septum and apex. Arrowheads indicate dyskinetic interventricular septal motion during systole. Minor-axis dimensions were obtained as shown, from which FS was calculated. LA = left atrium; LV = left ventricle.

determining outcome following LV aneurysmectomy? Second, among the patients with aneurysms that spare the base of the left ventricle, does fractional shortening (FS) provide an adequate assessment of residual function? Third, can this parameter of basilar LV function be accurately measured by means of routine 2DE techniques?

METHODS

Patients. Between January, 1981, and May, 1984, 44 patients underwent 2DE examination, 1 to 56 (mean 12.6) days prior to LV aneurysmectomy. Three patients had aneurysms involving the base of the left ventricle, three had technically unsatisfactory echocardiograms, and one had severe aortic insufficiency necessitating concurrent aortic valve replacement. These seven were eliminated from further evaluation, leaving a total of 37 patients. There were 33 men and 4 women, ranging in age from 36 to 75 (mean 58 + 10) years. The presence and location of the LV aneurysm were confirmed preoperatively by both angiographic and echocardiographic methods in all patients. The LV apex was involved in 34 patients. In one each, the aneurysmal segment was confined to the lateral, septal, or inferior walls.

Primary indications for surgery were intractable angina in 11 patients, congestive heart failure in 17, and refractory ventricular arrhythmias in nine. Seven of these latter nine patients also had significant symptoms of angina and/or congestive heart failure prior to aneurysmectomy. All patients for whom angina or congestive heart failure was the primary indication, and seven of nine patients referred to surgery for refractory arrhythmias, were in New York Heart Association (NYHA) functional class III or IV prior to surgery. Thirty-six of 37 patients had significant coronary artery disease at the time of cardiac catheterization, defined as >50% reduction in luminal diameter of one or more major epicardial vessels. One patient, with a discrete apical aneurysm, had angiograph- ically normal coronary arteries and underwent aneurysmectomy because of ventricular arrhythmias. Coronary artery revascularization was undertaken in 33 patients. In

no patient was there evidence of a significant change in clinical status during the interval between echocardiographic study and aneurysmectomy.

Follow-up. Follow-up data were available for all patients. Among survivors, duration of follow-up ranged from 10 to 39 (mean 17.9) months. On the basis of immediate surgical results and late follow-up, patients were allocated to one of two outcome groups (A or B). Criteria utilized to determine assignment to a group included survival and improvement or deterioration in NYHA functional classification. The persistence or resolution of objective evidence of ventricular tachyarrhythmias was not considered in assigning patients to an outcome group unless recurrent arrhythmias led to deterioration in functional class.

Cardiac catheterization. Left heart catheterization was performed by means of standard angiographic and hemodynamic techniques.14 LV pressures were measured by means of a calibrated fluid-filled catheter system and were recorded prior to contrast studies. Left ventriculography was performed in the right anterior oblique projection. Single-plane angiographic volumes were measured at end diastole and end systole according to the method of Sandler and Dodge.‘” Ejection fraction (EF) was calculated by the formula: EF = (end-diastolic volume - end- systolic volume)/end-diastolic volume. Basilar half EF was calculated according to the method of Kapelanski et al.*

2DE. 2DE was performed prior to LV aneurysmectomy by means of commercially available, wide-angle sector scanners with either a 3.0 or a 3.5 MHz mechanical transducer. The parasternal window was used to visualize the left ventricle in the long- and short-axis views. In the long-axis view, diastolic (Dd) and systolic (Ds) minor-axis dimensions at the base were measured. End diastole and end systole were defined as the frames just prior to mitral valve closure and mitral valve opening, respectively. These dimensions were aligned perpendicular to the long axis of the LV cavity at the level of the chordae tendineae. FS was calculated as: (Dd - Ds)/Dd (Figs. 1 and 2). In the short-axis view, diastolic (Ad) and systolic (As) areas were measured at the papillary muscle level. Fractional area

716 Ryan et al. April, 1986

American Heart Journal

Fig. 2. Group B. Parasternal long-axis views at end-diastole and end-systole from a patient who died soon after LV aneurysmectomy. Arrowheads indicate dyskinetic interventricular septal motion during systole. Basilar contractility is reduced as indicated by the reduced FS. Abbreviations as in Fig. 1.

Table 1. Clinical data

Clinical Group A Group H data (n = 27) (n = 10)

Male/female 2413 Y/l Age (yr) 57.6 * 10 58.9 i- 11 Recent myocardial infarction* 6 2 Preoperative functional class 3.6 +- 0.5 3.7 It 0.5 Indication for aneurysmectomy

Angina 8 3 Congestive heart failure 12 5 Ventricular tachyarrhythmia I 2

Age and preoperative functional class were expressed as mean t standard deviation. *Defined as occurrin(: within 8 weeks of aneurysmectomy.

change was calculated as: (Ad - As)/Ad. For these views, end diastole and end systole were defined as the frames depicting the greatest and least LV areas, respectively. Measurements were performed by means of a commercially available off-line analysis system.

In all cases, echocardiographic measurements were performed independently by two observers, one blinded to clinical outcome, the other nonblinded. At least three cardiac cycles were measured and the results averaged. Data reported henceforth are those obtained by the blinded observer. Results of the nonblinded observer were used to test interobserver reproducibility.

Statistics. Measurements on the two groups were compared by means of a two-tailed t test. Agreement between variables was assessed with the Pearson correlation coeffi- cient. The usefulness of FS to predict outcome was measured by sensitivity, percentage of poor outcomes with positive result, specificity, percentage of good outcomes with negative result, and predictive value, percentage of positive results with poor outcomes. Variables are expressed as mean f standard deviation.

RESULTS

Group A. Group A (n = 27) included all patients who clinically were believed to have benefited from resection of the aneurysm. All group A patients were

Table II. Angiographic and hemodynamic variables

Group Group A R

Number of diseased vessels 2.3 2 0.8 2.9 -+ 0.3* Number of bypass grafts 2.2 i 1.2 2.9 + 1.2 LVEDP (mm Hg) 20.5 5 8.0 26.0 + 9.0 Global EF 0.38 t 0.14 0.37 + 0.12 Basilar half EF 0.50 +- 0.09 0.37 f 0.10t

EF = ejection fraction; LVEDP = left ventricular end-diastolic pressure. Values are mean + standard deviation. *p < 0.05. tp < 0.001.

alive a mean of 18.2 (range 10 to 39) months after surgery. Among patients who underwent aneurysmectomy for symptoms of angina and/or congestive heart failure, assignment to group A required improvement by at least one functional class compared to preoperative status. Of the 20 patients who met this requirement, five improved to NYHA functional class I and 15 to functional class II. All had been considered functional class III or IV prior to aneurysmectomy.

Among patients referred to surgery for refractory ventricular tachyarrhythmias, only those who experienced improvement in clinical status were included in group A. Because seven of nine patients in this category had significant angina and/or congestive heart failure in addition to refractory arrhythmias, assignment to group A required improvement in these symptoms as well as resolution of the arrhythmias. Seven patients met these criteria and remained without evidence of recurrence during the follow-up period.

Group B. Group B consisted of patients who did not benefit clinically from aneurysmectomy. Includ- ed were 10 patients who either died during the follow-up period or who had no improvement in NYHA functional class compared to preoperative

"Olume 111

Number 4

.80

.60

. LO

p=NS p<O.OOl

1 1 I I

8 I I 1

A B A B

EF BHEF

.80

60

40

.20

Fig. 3. The angiographic indices of LV contractility as a function of outcome are shown. While ejection fraction (EF) fails to discriminate between patients in the two outcome groups (p = NS), basilar half EF (BHEF) is significantly greater among patients in group A (p < 0.001). Vertical bars represent mean t- standard deviation.

status. Five patients died within 7 days of surgery, yielding an overall perioperative mortality rate of 14%. This included two patients whose primary indications for surgery were ventricular arrhythmias. Preoperatively, both had severe LV dysfunction and both died as a result of pump failure, Three patients died between 5 and 7 months following aneurysmectomy. One died of respiratory failure secondary to recurrent pulmonary infections dating to the immediate postoperative period, one died of pump failure, and one died suddenly.

Two patients who underwent surgery for severe congestive heart failure remained in NYHA functional class IV after aneurysmectomy. Subsequent- ly, both have been hospitalized at least three times each for treatment of congestive heart failure.

Clinical variables. There was no significant difference between the two groups with respect to age, preoperative functional class, proximity to most recent myocardial infarction, or primary indication for surgery (Table I). Among eight patients who underwent surgery within 8 weeks of a documented myocardial infarction, six were in group A, one died during the immediate postoperative period, and one

.50

.40

.30

.20

.lO

2DE prior to aneurysmectomy 7 I 7

p<O.OOl p=NS

I 1 I I

.

.

. . l .

. .

I .

.

. . .

.

0 .

. . . . .

.

.

. . .

.

.

. . . . I .

.

I I I I A

.50

.40

.30

.20

.lO

A B A 5

FS FAC

Fig. 4. Individual values for 2DE parameters of LV systolic function are shown for each outcome group. FS provides good separation between patients with a good (group A) vs poor (group B) outcome (p < 0.001). Frac- tional area change (FAC) fails to separate the two groups (p = NS). Vertical bars represent mean + standard deviation.

remained in functional class IV. Seventy-eight per- cent of patients with ventricular tachycardia, 73% of patients with angina, and 71% of patients with congestive heart failure had a good clinical outcome.

Angiographic and hemodynamic variables. Angio- graphic and hemodynamic variables are summarized in Table II. Preoperative LV end-diastolic pressure ranged from 4 to 40 mm Hg. Among patients in group A, the mean was 20.5 f 8 mm Hg vs 26.0 f 9 mm Hg among patients in group B 0, = NS). Twen- ty-two patients had triple-vessel disease, 11 had two-vessel disease, and three had single-vessel disease. In one patient, no significant coronary artery narrowing Gas observed. The number of diseased vessels was greater among patients in group B than in group A (p < 0.05). Between one and five (mean 2.5) coronary artery bypass grafts were placed in 33 patients. There was no significant relationship between the number of bypass grafts and clinical outcome.

EF, as measured from right anterior oblique



Table III. 2DE variables ---

Group Group

A A

Dd (cm) range 3.2 - 6.5 3.9 - 6.6 mean 4.9 -+ 0.8 5.2 k 0.7

Ds (cm) range 2.4 - 5.3 3.7 - 5.9 mean 3.6 t 0.8 4.4 + 0.6*

FS 0.25 I 0.06 0.15 * 0.04t FAC 0.31 -+ 0.10 0.26 zk 0.08

Abbreviations: Dd = end-diastolic minor-axis dimension at the base; 1)s = end-systolic minor-axis dimension at the base; FS = fractional shortening: FAC = fractional area change. FS and FAC were calculated as defined in the text; values are mean f standard deviation. *p = 0.02. tp < 0.001.

angiograms, ranged from 0.19 to 0.74 (Fig. 3). Mean values for patients in group A vs group B were 0.38 k 0.14 and 0.37 + 0.12, respectively. There was no significant correlation (p = NS) between preoperative global EF and postoperative functional class. Basilar half EF was significantly greater (p < 0.001) among patients in group A (0.50 + 0.09) than Group B patients (0.37 + 0.10).

2DE variables. LV minor-axis dimensions at the base of the heart were obtained in all 37 patients. Because of the quality of the image, mid-LV area determinations were performed in only 27 (73%). The problem of lateral resolution, an issue when planimetering short-axis areas, but not when measuring minor-axis dimensions, was the primary rea- son for the difference in success rates. These data are summarized in Table III.

End-diastolic minor-axis dimension and fractional area change were poor predictors of surgical outcome. Mean end-systolic minor-axis dimension was significantly smaller in group A (3.6 + 0.8 cm) than in Group B (4.4 + 0.6 cm, p = 0.02). Despite the statistical significance, considerable overlap existed between the two groups.

FS provided better separation than minor-axis dimension (Fig. 4). All seven patients with FS < 0.17 fell into group B, while 25 of 27 patients (93%) with FS > 0.17 were in group A (Table IV). Three patients had a FS = 0.17, two in group A and one in group B. With the use of a value of <0.17 to identify patients likely to experience a poor outcome, sensitivity was 70%, specificity 100%) and predictive value 100%.

FS as a predictor of residual myocardial function. Both FS and basilar half EF provided excellent separation between groups. Similar degrees of overlap existed with both measurements. To test wheth- er FS parallels basilar half EF as a means of

Table IV. Relationship of preoperative FS to outcome following aneurysmectomy

Outcome

Group A Group H

Fractional 20.17 27 3 Shortening <0.17 0 7

assessing residual myocardial function, linear correlation was performed. A high correlation existed (r = 0.84), indicating that the two parameters are similar in their ability to assess basilar systolic myocardial function.

Reproducibility. Interobserver reproducibility for the determination of FS was excellent. Mean difference between values was 0.02 and in only 3 of 37 cases (8%) did observers disagree by >0.06.

DISCUSSION

The primary goal of this investigation was to identify a subset of patients who, on the basis of preoperative assessment, would be unlikely to benefit from aneurysmectomy. By means of retrospective analysis of surgically treated patients, this study identified a group of patients with FS < 0.17 who experienced a poor outcome. These results demonstrate that echocardiographic assessment of basilar LV function is useful in the management of patients with aneurysms that spare the base of the heart.

An important aspect of this study is the inclusion of patients in whom ventricular arrhythmias were the primary indication for aneurysmectomy. Refrac- tory arrhythmias are a relatively common indication for this surgical procedure.“j Clearly, outcome of these patients is heavily dependent on the success with which the arrhythmia can be controlled surgically. An assessment of residual LV function is also important. The majority of patients in this subset have significant underlying LV dysfunction.17* I8 In one investigation,‘? in which 49 patients underwent LV aneurysmectomy for treatment of life-threaten- ing ventricular arrhythmias, most late deaths were the result of congestive heart failure. Our data support these findings. Among nine patients referred to surgery for intractable arrhythmias, seven experienced a good surgical result and, with evidence of adequate preoperative LV reserve, remained alive and clinically improved after a mean follow-up of 16.9 months. Two patients died as a result of pump failure. Both had very poor preoperative basilar LV function as measured by both basilar half EF and echocardiographic FS.

2DE prior to aneurysmectomy 719

As successful surgical treatment of ventricular arrhythmias increases, a growing number of patients with LV aneurysms are being considered candidates for this mode of therapy. Thorough preoperative evaluation of residual LV function appears warrant- ed. Specifically, a need exists to identify patients who, despite effective surgical ablation of the arrhythmogenic focus, have inadequate LV myocardial reserve to permit a successful outcome.

The importance of measuring contractility of the nonaneurysmal segment and its influence on both operative and long-term survival rates has been demonstrated by several investigators. Watson et al.,7 by means of a double hemispheroidal model, derived contractile segment EF from single-plane angiograms. They reported functional improvement and a higher survival rate among patients with a contractile segment EF > 0.45. Absolute volume of the contractile segment, however, did not influence outcome. Kiefer et al.,s by means of a similar approach, demonstrated that survival and improved functional class were most likely in patients with a contractile segment EF > 0.35 and LV end-diastolic pressure <25 mm Hg. Kapelanski et al.8 measured contractile segment and basilar half EF in both single and biplane angiographic projections. Basilar half EF was superior to both contractile segment and global EF in predicting outcome. This relationship was particularly helpful in patients undergoing urgent surgery. We obtained similar results. In our series, a basilar half EF > 0.37 was present in all patients in group A but in only two patients in group B.

2DE has certain advantages over angiographic techniques for the assessment of basilar LV function. It is noninvasive and provides a resting, non- stressed measure of systolic function. The use of contrast agents in patients with significant LV dysfunction may be hazardous and, by depressing myocardial contractility, can lead to an underesti- mation of true systolic function. Multiple views of the left ventricle can be recorded during the echocardiographic examination, permitting the extent and location of the aneurysm to be accurately delineated.

2DE can also be used to evaluate the nonaneurys- ma1 myocardium. Barrett et all9 derived an index of residual myocardium by means of the 2DE equiva- lent of the right anterior oblique angiogram. This index, which correlated well with 6 months’ survival in medically treated patients, measured the amount but not the contractility of the nonaneurysmal LV myocardium. We obtained similar results, suggest- ing that, while global EF was a poor predictor of

survival and improved functional class, basilar half EF correlated well with outcome. Thus our findings are in agreement with those in which the importance of quantitating residual myocardial function in the management of patients with LV aneurysms was demonstrated.

The use of FS as an index of LV contractility is based on angiographic data which demonstrate that shortening of the minor axis is the primary contrib- utor to LV stroke volumezo and the percentage of change in minor-axis dimension is linearly correlated with EF.21 Echocardiographic FS is a readily obtained parameter which reflects dimensional changes along a single axis at the base of the left ventricle. To imply that FS depicts residual myocardial function in patients with LV aneurysms entails several assumptions. First, the aneurysm must not involve that portion of the LV in which FS is measured. Since minor-axis dimensions at the base are used to derive FS, only in patients with aneurysms sparing the base can one assume that FS reflects residual myocardial function. One must further assume that linear changes along a single axis are representative of regional contractility. Our results, that basilar half EF and FS were well correlated (F = 0.84), support this assumption. Angiographic21 as well as M-mode echocardiographic2* studies suggest that FS correlates well with global LV function in patients without regional wall motion abnormali- ties. Thus FS should reflect regional function at the base, provided that normal basal wall motion is preserved and areas of dysfunction are confined to more apical regions. Similar geometric assumptions must be made when single-plane angiograms are used to derive basilar half volumes and EF.8 A final potential problem, common to all ejection phase indices, is that they are dependent on both pre- and afterload. Thus changes in end-diastolic volume or blood pressure may alter FS independent of any change in contractility.

FS has traditionally been derived from minor-axis dimensions at the chordal level by means of M-mode echocardiography. 23,24 Although this represents a standardized, reproducible measurement,25 there are potential sources of error which stem primarily from the lack of spatial orientation inherent in M-mode echocardiograms. Specifically, if the M-mode beam transects the left ventricle obliquely, significant overestimation of dimensions will occur, resulting in artifactually low FS. This is particularly likely in patients with LV aneurysms because of distortion of LV shape and position. 2DE overcomes this problem by providing accurate spatial orientation. By align- ing the measurements perpendicular to the long-axis



of the LV cavity, true minor-axis dimensions can be recorded reliably-l3

A major limitation of this study is the bias introduced by exclusion of medically treated patients with LV aneurysms. Indications for LV aneurysmectomy are highly variable.16 Most patients who do not undergo surgery fall at the two extremes of LV function. Either they respond to medical therapy and are relatively well compensated or they have markedly depressed LV function and are considered unsuitable surgical candidates. Exclusion of these patients could bias our results. However, patients entered into this study exhibited a wide range of ventricular function (as measured by angiographic EF) and, on the basis of clinical, hemodynamic, and angiographic parameters, appear to be fairly representative of all patients with LV aneurysms.

In conclusion, the role of quantitative 2DE in the evaluation of patients undergoing LV aneurysmectomy was examined. FS is easily obtained from 2DE and provides a reproducible measure of basilar LV function. In patients with aneurysms that spare this portion of the left ventricle, FS adequately reflects the level of contractility of the nonaneurysmal myocardium. It is as accurate as basilar half EF in predicting outcome following LV aneurysmectomy and is superior to parameters which measure global LV function. Our results suggest that 2DE may be used to identify patients least likely to benefit from surgical intervention.

We wish to acknowledge the statistical assistance of Naomi Fineberg, Ph.D., and the secretarial support of MS Nancy Naan.

REFERENCES

1. Aranda JM, Befeler B, Thurur R, Vargas A, El-Sherif N, Lazzara R: Long-term clinical and hemodynamic studies after ventricular aneurysmectomy and aorta-coronary bvnass. 3 Thorac Cardiovasc Surg 73:772, 1977.

2. Frbelich RT, Falsetti HL, Doty DI3, Marcus ML: prospective study of surgery for left ventricular aneurysm. Am J Cardiol 45:92-l, 1980.

3. Lee D C-S, Johnson RA, Boucher CA, Wexler LF, McEnany MT: Angiographic predictors of survival following left ventricular aneurysmectomy. Circulation 56: (Suppl IIl:II-128, 1977.

4. Cosgrove DM, Loop FD, Irarrazaval MJ, Groves LK, Taylor PC, Golding LA: Determinants of long-term survival after ventricular aneurysmectomy. Ann Thorac Surg 26:357, 1978.

5. Burton NA, Stinson EB, Oyer PE, Shumway NE: Left ventricular aneurysm. Preoperative risk factors and long- term postoperative results. J Thorac Cardiovasc Surg 77:65, 1979.

6. Loop FD, Cosgrove DM: Results of ventricular aneurysmectomy. Am d Surg 141:684, 1981.

7. Watson LE, Dickhaus DW, Martin RH: Left ventricular aneurysm. Preopertive hemodynamics, chamber volume, and results of aneurysmectomy. Circulation 52:868, 1975.

8. Kapelanski DP, Al-Sadir J. Lamberti JJ, Anagnostopoulos CE: Ventriculographic features predictive of surgical outcome for left ventricular aneurysm. Circulation 561167, 1978.

9. Kiefer SK, Flaker GC, Martin RH, Curtis JJ: Clinical improvement after ventricular aneurysm repair: Prediction by angiographic and hemodynamic variables. J Am Call Cardiol 2:30, 1983.

10. Dillon JC. Feigenbaum H, Weyman AE, Corya BC, Peskoe S, Chang S: M-mode echocardiography in the evaluation of patients for aneurysmectomy. Circulation 53:657, 1976.

11. Weyman AE, Peskoe SM, Williams ES, Dillon JC, Feigen- baum H: Detection of left ventricular aneurysms by cross- sectional echocardiography. Circulation 54:936, 1976.

12. Rakowski H, Martin RP, Schapira JN, Wexler L, Silverman JF, Cipriano PR, Guthaner DF, Popp RL: Left ventricular aneurysm; Detection and determination of resectibility by two-dimensional echocardiographv (Abstrl. Circulation 56:

- (suppl III):III-153, 1977. 13. Chazal RA. Armstrone WF. Finebere NS. Green D. Dillon JC.

Feigenbaum H: Critical evaluationof left ventricular minor axis dimensions using two-dimensional echocardiography: Reproducibility and relationship to M-mode measurements (abstr). Clin Res 31:174A, 1983.

14. Grossman W: Cardiac catheterization and angiography, ed 2. Philadelphia, 1980, Lea & Febiger, Publishers, pp 37-211.

15. Sandier H, Dodge HT: The use of single plane angiocardio- grams for the calculation of left ventricular volume in man. AM HEART J 75:325, 1968.

16. Cohen M, Packer M, Gorlin R: Indications for left ventricular aneurysmectomy. Circulation 67:717, 1983.

17. Buda AJ, Stinson EB, Harrison DC: Surgery for life-threat- ening ventricular tachyarrhythmias. Am J Cardiol 44:1171, 1979.

18. Harken AH, Horowitz LN, Josephson ME: Comparison ot standard aneurysmectomy and aneurysmectomy with direct- ed endocardial resection for the treatment of recurrent sustained ventricular tachycardia. J Thorac Cardiovasc Surg 60~527. 1980.

19. Barrett MJ, Charuzi Y, Corday E: Ventricular aneurysm: Cross-sectional echocardiographic approach. Am J Cardiol 46:1133, 1980.

20. Hood WP, Jr Rolett EL: Patterns of contraction in the human left ventricle (abstrl. Circulation 4O:(suppl III):III- 109, 1969.

21. Lewis RP, Sandler H: Relationship between changes in left ventricular dimensions and the ejection fraction in man. Circulation 44:548, 1971.

22. Quinones MA, Pickering E, Alexander JK: Percentage of shortening of the echocardiographic left ventricular dimension. Its use in determining ejection fraction and stroke volume. Chest 74:59, 1978.

23. McDonald IG, Feigenbaum H, Chang S: Analysis of left ventricular wall motion by reflected ultrasound. Application to assessment of myocardial function. Circulation 46:14, 1972.

24. Gutgesell HP, Paquet M, Duff DF, McNamara DG: Evalua- tion of left ventricular size and function by echocardiography. Results in normal children. Circulation 56:457, 1977:

25. Crawford MH. Grant D. O’Rourke RA. Starling MR. Groves BM: Accuracy and reproducibility of new M-mode kchocar- diographic recommendations for measuring left ventricular dimensions. Circulation 61:137, 1980.

quantitative two-dimensional echocardiographic assessment of patients undergoing left ventricular...

Documents