International Journal of Cardiology, 37 (1992) 351-359 0 1992 Elsevier Science Publishers B.V. Ail rights reserved 0167-5273/92/$05.00

351

CARD10 01549

Early diastolic left ventricular function as a marker of acute cardiac rejection: a prospective serial echocardiographic study

Jai-Wun Park ‘, Henning Warnecke b, Mario Deng a, Stefan Schiiler b, Karl Wilhelm Heinrich a and Roland Hetzer b

a Herzzentrum Duisburg KWX; Duisburg, Germany, and b Deutsches Herzzentrum Berlin, Berlin, Germany

(Received 25 February 1992; revision accepted 16 June 1992)

Park J-W, Warnecke H, Deng M, Schiiler S, Heinrich KW, Hetzer R. Early diastolic left ventricular function as a marker of acute cardiac rejection: a prospective serial echo-cardiographic study. Int J Cardiol 1992;37:351-359.

Changes in left ventricular early diastolic time intervals are sensitive indicators of incipient left ventricular dysfunction. We tested the hypothesis that acute rejection in cardiac transplant recipients is associated with alteration of early diastolic myocardial function, as expressed by the time interval Te, a parameter derived from digitized M-mode echocardiograms. Te is defined as the time interval between maximal posterior wall contraction and the point of peak posterior wall endocardium retraction velocity, as determined by the nadir of the computed first derivative curve. In transplant patients without rejection (group A, IZ = 481, Te was prolonged compared to healthy individuals (group C, n = 35) (79.0 + 12.5 ms vs 64.0 & 7.9 ms; p < 0.0001). During acute rejection (group B, n = 18) transplant patients had significantly longer mean Te values compared to transplant patients without rejection (group A) (97.8 f 17.9 ms vs 79.0 k 12.5 ms; p < 0.0001). Longitudinal studies in individual patients (group D, n = 18) demonstrated that rejection is associated with prolongation of Te (94.5 i 16.0 ms during rejection vs 79.0 ? 10.3 ms before rejection; p < 0.0002) and that Te returns to individual baseline values in response to treatment (79.2 k 9.4 ms after therapy vs 79.0 ?I 10.3 ms before rejection; NS). In a prospective study, Te changes in transplant patients (group E, n = 96) were correlated with myocardial biopsy results. Sixty-one biopsies showed acute rejection, and 115 biopsies were negative. Taking a 20% increase in Te compared to pre-rejection values as the criterion for an acute rejection, the sensitivity, specificity, positive and negative predictive values were 80, 94, 88, and 90%, respectively. We conclude that the parameter Te is a useful non-invasive marker of acute cardiac rejection.

Key words: Time interval Te; Myocardial biopsy; Allograft rejection

Introduction

Correspondence tot Dr. J.-W. Park, Herzzentrum Duisburg KWK. Gerrickstr. 21. 4100 Duisburg 12, Germany. Tel. 0203- 4508-O.

Since the introduction of cyclosporin in the routine management of heart transplantation,

352

survival has improved, but early detection of acute cardiac allograft rejection has become more diffi- cult [l]. Myocardial biopsy, considered to be the gold standard despite certain limitations, is in- convenient to the patient, carries a minor risk of complications inherent to an invasive procedure, and leaves diagnostic gaps in long-term surveil- lance. In order to provide non-invasive diagnostic alternatives to myocardial biopsy, a variety of techniques, among which several echocardio- graphic methods, have been proposed [2-121. Un- til this time, each procedure has served only as a screening method, supplementing myocardial biopsy rather than being able to replace it. We examined an early diastolic time interval parame- ter in search of a sensitive and reliable echocar- diographic marker of acute rejection, as it is known that diastolic parameters are sensitive in- dicators of failing left ventricular function in sev- eral cardiac diseases [13-161. Early diastolic pa- rameters can easily be derived from digitized M-mode echocardiograms and are consistently reproducible [17-201. The time interval parame- ter Te [21] is believed to reflect the myocardial relaxation process and to be relatively indepen- dent of loading conditions [22,23]. This study was designed to evaluate the use of the parameter Te as a marker of acute cardiac allograft rejection.

Materials and Methods

2. Is there a significant reversible prolongation of Te in individual transplant patients during acute rejection (study 2)?

3. What is the clinical applicability of the pro- posed parameter in rejection monitoring ex- pressed by its prospectively determined sensi- tivity and specificity (study 3)?

The three studies were conducted consecutively.

Patient population

All transplant patients studied were on a triple immunosuppressive protocol including cyclo- sporin (2-5 mg/kg/day), azathioprine (l-3 mg/kg/day) and prednisolone (0.15 mg/kg/day). In addition, they received furosemide and capto- pril according to blood pressure levels.

Study 1

We compared 48 cardiac allograft recipients with microscopically excluded rejection (group A: 42 males, 6 females; 5 to 63 yr, mean age 44.8 f 14.5 yr; 17 days to 3.6 yr, mean 337 days after cardiac transplantation) with 16 cardiac allograft recipients with moderate rejection according to the Billingham classification (group B: 14 males, 2 females; 18 to 64 yr, mean age 42.9 + 13.2 yr; 19 days to 3.5 yr, mean 319 days after cardiac trans- plantation) and 35 healthy volunteers (group C: 18 to 66 yr, mean age 40.9 f 14.8 yr).

Study design Study 2

We tested the following three hypotheses. 1. There is a difference in average Te values

between transplant patients as a group and normals.

2. There is a reversible prolongation of Te during acute rejection in the individual course.

3. The parameter Te is a sensitive and specific tool for non-invasive rejection monitoring. We conducted a three-phase study with the

following questions. 1. Are there different group mean Te values

between transplant patients in a non-rejecting state as documented by endomyocardial biopsy, transplant patients in a rejecting state and healthy persons (study l>?

In order to study individual rejection episodes, 18 transplant patients (group D: 14 males, 4 fe- males; 22 to 59 yr, mean age 42.5 + 10.2 yr) were consecutively examined by echocardiography, be- fore, during, and after successful treatment of acute rejection. The pre-rejection value of Te was calculated as a mean of at least two consecutive measurements.

Study 3

In order to calculate sensitivity, specificity, positive and negative predictive values of the proposed echocardiographic method, 96 trans-

353

plant patients (group E: 84 males, 12 females; 24 to 61 yr, mean age 43.3 k 10.5 yr> were prospec- tively studied between 1 April and 31 May 1989. We excluded the Te values during the first 14 days from the study because the immediate, post-transplantation Te values were observed to be significantly prolonged, but with a steady de- cline within 2 weeks. During this time there was no microscopic evidence of rejection in many cases.

Examination

The echocardiographic examinations were per- formed within 24 h before myocardial biopsy by an experienced echocardiographer using the Toshiba Sonolayer SSH 160A equipment and 2.5 MHz probes. The examiner was blinded to the patients’ clinical history and biopsy findings. The

M-mode echo

l End PW

patients rested for 15 min before the examination in order to attain a steady-state basal heart rate and blood pressure. The procedure was per- formed with the patient lying in the left lateral decubitus position. M-mode recordings were per- formed with simultaneous two-dimensional echocardiographic control and ECG recording, the cursor positioned at the level of the free edges of the mitral valve leaflets. Strip chart recording paper speed was 100 mm/s. All pa- tients were in sinus rhythm. Premature atria1 and ventricular complexes were excluded from analy- sis. Heart cycles with recipient atria1 contraction in late systole were not excluded. Only those recordings were considered adequate for further computer analysis in which septal and posterior wall echo-lines could clearly be identified for at least three cardiac cycles. Two independent ex- aminers digitized the echocardiograms by means

lo2

119 T HH End PW

0.40 -

TIME set

0.00 - >

0.00 0. 42 084

102

_ HH _ -s-e?- - dEndPW/dt

2.53

E TIME

0 00 0.42 0 8’4

Fig. 1. Left ventricular posterior wall endocardium of an original M-mode tracing (left), after digitizing the posterior wail

endocardium (right top) and calculating the corresponding first derivative curve (right bottom). End PW = posterior wall

endocardium; 0 = echocardiographic point of end-systole; E = point of peak negative dEnd/dt (peak endocardial retraction velocity) Te = time interval between 0 and E.

354

of a hand controlled cross-wire cursor and were processed by an IBM AT 02 computer. The de- tails of this method are described elsewere [24].

Echocardiographic parameters

The first derivative of the posterior wall endo- cardial motion was computed in order to calcu- late instantaneous motion velocity (Fig. 1). The parameter Te is defined as the time interval (ms) between end-systole (first derivative curve = 0) and the point of peak endocardial retraction ve- locity E [25]. The value of Te was obtained by averaging three consecutive measurements.

Beat-to-beat variability (10 consecutive meas- urements in each of 5 patients without and dur- ing acute rejection), day-to-day variability (2 measurements in 10 patients without acute rejec- tion 2 weeks apart), interobserver variability (10 echo-strips digitized by 2 readers independently) and intraobserver variability (10 echo strips digi- tized by 1 observer 2 weeks apart) were assessed.

Since early diastolic parameters of ventricular function are influenced by heart rate [261, we studied the relationship between heart rate and Te. In 5 patients and 5 controls atria1 pacing (donor atria1 pacing in transplant recipients) was performed (increase in heart rate by lO/min up to the Wenckebach point, 15 s pacing duration, 5 min rest between two pacing periods) and Te simultaneously determined.

Calculation of standard M-mode parameters (left ventricular end-diastolic .diameter, left ven- tricular end-systolic diameter, fractional shorten- ing, interventricular septum diameter, posterior wall diameter) was performed according to the recommendations of the American Society of Echocardiography [27].

Myocardial biopsy

Myocardial biopsies were, interpreted by an experienced pathologist. The myocardial biopsy was considered positive when moderate or severe rejection according to the Billingham classifica- tion [28] was present.‘The first control biopsy and corresponding Te value following acute rejection treatment were excluded from the study.

Statistical analysis

The results are presented as mean + SD. The unpaired Student’s t-test was performed to evalu- ate differences between groups (study 1). A p value < 0.05 was considered to be statistically significant. Intraobserver, interobserver, and day- to-day variability were expressed as a mean per- cent error (difference/average x loo), beat-to- beat variability was expressed as a variation coef- ficient.

For the statistical evaluation of individual fol- low-ups (study 21, the paired Student t-test was used.

In order to calculate sensitivity, specificity, positive and negative predictive values, acute re- jection was echocardiographically defined as an increase in Te > 20%, this criterion being prospectively tested, compared to the mean of 2 successive measurements in the non-rejecting state. An increase of 20% was arbitrarily chosen as a threshold value, as it exceeded the sum of the measured variabilities (biological, day-to-day, intra-observer).

Results

Reproducibility

The technical variability of the digitizing pro- cedure and the biological variability (beat-to-beat,

TABLE 1

Reproducibility of Te

Technical variability of the digitizing procedure (%) intraobserver variability

(10 echo strips) 3.1

interobserver variability

(10 echo strips) 3.0

Biological variability (o/o) day-to-day variability

(10 transplant patients,

examination interval > 14 days) 9.6

beat-to-beat variability

(10 consecutive beats)

5 transplant patients

without rejection 5.6

5 transplant patients during acute rejection 6.9

355

day-to-day) were approximately 3% and < lo%, respectively, yielding a good reproducibility of Te (Table 1).

Heart rate dependence

An inverse relationship between Te and heart rate was found during a stepwise increase in heart rate by atria1 pacing in transplant patients as well as in the control group (Fig. 2).

Study 1 (comparison of groups)

Transplant patients without acute rejection (group A) had significantly longer Te values than healthy controls (group C) (79.0 f 12.5 ms vs 64.0 &- 7.9 ms; p < 0.0001) (Fig. 31, significantly higher heart rates (100.7 * 17.3/min vs 68.9 L- 10.4/min;

80

70

60

50

40 1 60 70 80 90 100 110 120 130 l40 150 160 170 heart rate

Ibcatslmln)

transp1mt

TC (msecl

120

110

100

90

80

70

60

50

rn

patients

I

+

-*

. +

. + 9

A’ . A

60 70 80 90 100 110 120 130 140 150 160 170 heart rate (beatslminl

Fig. 2. Relationship between TE and heart rate in 5 normals

(top) and 5 transplant patients (bottom) during (donor-) atria1

pacing.

controls n-35

TX-pments TX-pahents

16ARnMB +AhnMB n=l+B n=16

t

ii

I

.

.

I + p~oooo1l+P~~~1----I

Fig. 3. Te values of healthy controls, a group of transplant

patients without acute cardiac rejection and a group of trans-

plant patients during acute cardiac rejection. TX-patients =

transplant patients; AR = acute rejection; MB = myocardial

biopsy.

p < 0.0001) and significantly higher systemic arte- rial blood pressure levels than normals (146.4 I!I l&6/100.9 f 17.5 mmHg vs 120.6 f 12.2/77.1 -t 9.2 mmHg; p < 0.0001). The group mean Te value of transplant patients during acute rejection (group B) was significantly longer than a group mean Te value in non-rejecting periods (group A) (97.8 k 17.9 ms vs 79.0 &- 12.5 ms; p < 0.0001) (Fig. 3).

Study 2 (individual courses)

Follow-ups in 18 individual patients (group D) showed a significant increase in Te during micro- scopically documented acute rejection (94.5 f 16.0 ms during acute rejection vs 79.0 _t 10.3 ms before acute rejection; p < 0.0002) (Fig. 4). After successful treatment, Te returned to values of the non-rejecting state (79.2 k 9.4 ms after therapy vs 94.5 f 16.0 ms during acute rejection; p < 0.0003) so that Te values after successful treatment and before rejection were comparable (79.2 f 9.4 ms after therapy vs 79.0 & 10.3 ms before rejection; NS). Heart rate (100.6 IL- 18.5/min vs 99.0 IL 15.3/min), blood pressure (145 f 16.7/99 k 16.3 vs 147 k 18.4/102 k 19.3 mmHg) and left ventric-

3.56

64 ms 119 ms

Fig. 4. M-mode echocardiograms from a transplant patient without (left) and during (right) acute cardiac rejection showing the

rejection-related prolongation of the early diastolic time interval parameter Te derived from the posterior wall endocardium.

ular end-diastolic diameter (45.1 + 6.1 vs 45.5 + septum diameter (11.6 + 1.9 vs 12.4 k 2.1 mm) 5.6 mm), left ventricular end-systolic diameter and posterior wall diameter (10.7 k 2.1 vs 11.3 f (31.8 + 5.7 vs 31.5 + 5.4 mm), fractional shorten- 1.8 mm) during acute rejection did not differ ing (29.5 f 7.1 vs 30.8 f 6.8%), interventricular significantly from non-rejecting periods.

TABLE 2

Sensitivity, specificity, and predictive values of the echocar- diographic parameter Te in rejection diagnostics using the threshold of 20% increase.

Echo

BX : w

Sensitivity: 80%; specificity: 94%; positive predictive value: 88%; negative predictive value: 90%; BX = endomyocardial biopsy.

Study 3 (prospective study)

Choosing a 20% increase in Te compared to the non-rejecting state Te value as the criterion for the diagnosis of an acute allograft rejection, sensitivity and specificity were calculated to be 80% and 94%, respectively. Positive and negative predictive values were 88% and 90%, respectively (Table 2).

Discussion

Despite certain limitations, myocardial biopsy is currently considered to be the diagnostic gold standard in the detection of an acute allograft rejection. Its proposed replacement by contempo- rary non-invasive methods has not gained much confidence. The implementation of a non-inva- sive, sensitive, reproducible and easy-to-handle marker of acute rejection remains a continuing challenge.

Rationale of the parameter Te

As it is known that impairment of early dias- tolic relaxation constitutes one of the first mani- festations of mechanical dysfunction in the hu- man left ventricle in diverse diseases [13- 16,32,331, we searched for an echocardiographic parameter capable of sensitively and reliably indi- cating alterations of intrinsic myocardial relax- ation properties. Wall thickness based parame-

ters are known to carry a significant amount of variability [27] apt to obscure wall dimension changes induced by acute rejection. Left ventricu- lar dimensional changes cannot be assessed reli- ably, since septal wall motion is known to be disturbed postoperatively [34,35] and septal and posterior wall motion are not in phase [36]. Digi- tized M-mode measurements of early diastolic time intervals have been shown to be repro- ducible [17-20,22,23]. It seemed reasonable to analyse a well-defined part of the posterior wall endocardium. Given an orthogonal cursor ap- proach, this area yields a well identifiable and reproducible endocardium contour detection. The endocardium contour-derived time-interval pa- rameter Te, which reflects left ventricular early diastolic relaxation properties, has been proposed to be relatively independent of pre-load condi- tions [25].

Critical evaluation of the results

The results of studies 1 and 2 indicate that heart transplant recipients in a non-rejecting state have a slower early diastolic relaxation compared to normal controls, as expressed by the parameter Te, and show further prolongation of the early diastolic relaxation process during acute rejection compared to the transplant patient group without rejection. The results of these studies encouraged us to perform the prospective serial echocardio- graphic study 3 to evaluate the clinical usefulness of the proposed parameter Te in managing trans- plant patients. The calculated sensitivity of 80% and specificity of 92% do not, in our opinion, justify the abolishment of endomyocardial biopsy but the institution of an altered time schedule.

The prolongation of Te in non-rejecting trans- plant patients may reflect a reduced left ventricu- lar compliance due to wall hypertrophy following cyclosporin-induced hypertension, cyclosporin- and rejection-related fibrosis and transient is- chemia in the perioperative period [8]. The addi- tional increase of Te during acute rejection can be explained by the development of edema and cellular infiltrates, suggesting a reversible restric- tive hemodynamic pattern which has been de- scribed earlier [3].

358

There are factors other than relaxation prop- erties capable of influencing Te. Alterations of preload, especially the timing of recipient atria1 contraction which contribute substantially to the variability of Doppler echocardiographic parame- ters such as pressure half-time [34], peak early mitral flow velocity, and isovolumic relaxation time seem to exert a less pronounced influence on Te. The beat-to-beat variability of consecutive cardiac cycles was below 7%. The variabilities in general seemed to be markedly smaller than the Te value increase due to acute rejection. As it is known that heart rate has a’i important influence on early diastolic parameters of left ventricular function [26], we examined the relationship be- tween Te and heart rate and found it clearly to be inverse, but less pronounced, in transplant patients. Although there is a reduced heart rate modulation in the transplanted heart as ex- pressed by comparable heart rates in the reject- ing and non-rejection state, this might account for 5 of 12 cases with false negative echocardio- graphic results in this study in which there was a significant increase in heart rate during acute rejection (> 20/min). Afterload and systolic per- formance, which remained virtually unaltered during acute rejection, represent further determi- nants. Aging and time post transplantation may have added to the variability of Te, an issue which should be addressed in follow-up studies. Furthermore, the parameter Te, representing a regio’nal assessment of posterior wall endo- cardium, might be insensitive in asymmetric pat- terns of acute rejection which have been reported [35]. Furthermore, every process causing myocar- dial damage otlier than rejection could explain false positive Te values.

A prolongation of Te without histological signs of rejection may, of course, be explained by a false negative biopsy result Which - according to our experience - is most frequently caused by a sampling error. Furthermore, it is conceivable, that in humorally mediated rejection - without microscopic evidence of interstitial mononuclear infiltrates and myocyte necrosis - vascular alter- ations contribute substantially to alteration of functional parameters during acute rejection [36,371. This eventually leads to the question of

whether rejection treatment should be imple- mented on the basis of altered functional param- eters of acute rejection or on the basis of biopsy results. Endomyocardial biopsy yields histological information on myocardial specimens but no functional data. Implementation of rejection treatment is based on the idea that this histologi- cal information coincides with functional alter- ations in an undetectable early stage. The intro- duction of parameters which reproducibly and validly detect early functional alterations seems to allow treatment implementation on a sound basis. It has been our experience, that treatment can be delayed without risk for the patient, when these functional parameters are completely nor- mal, even in the presence of histologically moder- ate rejection.

Conclusion

Acute cardiac rejection in the cyclosporin- treated cardiac allograft recipient is associated with a significant increase in the early diastolic relaxation parameter Te which can easily be de- rived from a digitized M-mode echocardiogram. Te returns to individual baseline values after implementation of successful treatment. There- fore Te can be used as a non-invasive method for rejection monitoring.

References

Oyer PE, Stinson EB, Jamieson SW et al. Cyclosporine in cardiac transplantation: a 2f year follow-up. Transplant Proc 1983;15:2546-2552. Popp RL, Schroeder JS, Stinson EB, Shumway NE, Harri- son DC. Ultrasonic studies for the early detection of acute carbiac rejection. Transplantation 1971;11:543-550. Sagar KB, Hastillo A, Wolfgang TC, Lower RR, Hess ML. Left ventricular mass by M-mode echocardiography in cardiac transplant patients with acute rejection. Circula- tion 1981;64(suppl 11):216-220. Nowygrod R, Spotnitz HM, Dubroff JM, Jardy MA, Reemtsma K. Organ mass: an indicator of heart transplant rejection. Transplant Proc 1983;15:1225-1228. Dawkins KD, Oldershaw PJ, Billingham ME et al. Nonin- vasive assessment of cardiac allograft rejection. Transplant Proc 1985;17:215-217. Dubroff JM, Clark MB, Wong CYH et al. Changes in left ventricular mass associated with the onset of acute rejec- tion after cardiac transplantation. Heart Transplant 1984;3:105-109.

359

7 Paulsen W, Magid N, Sagar K et al. Left ventricular

function of heart allografts during acute rejection: an

echocardiographic assessment. Heart Transplant 1985;

4525-529.

8 Valantine HA, Fowler MB, Hunt SA et al. Changes in

Doppler echocardiographic indexes of left ventricular

function as potential markers of acute cardiac rejection.

Circulation 1987;76(suppl V):86-92.

9 Desruennes M, Corcos T, Cabrol A et al. Doppler

echocardiography for the diagnosis of acute cardiac allo-

graft rejection. J Am Co11 Cardiol 1988;12:63-69.

10 Angermann CE, Spes CH. Hart RJ, Kemkes BM, Gokel

MJ, Theisen K. Echocardiographic diagnosis of acute re-

jection in heart transplant recipients on cyclosporine. Z

Kardiol 1989:78:243-252.

11 Ciliberto GR, Cataldo G, Cipriani M et al. Echocardio-

graphic assessment of cardiac allograft rejection. Eur Heart

J 1989;10:400-408.

12 Park JW, Schiiler S. Urbanczyk M, Scheller E, Hetzer R,

Fleck E. LV early diastolic relaxation by M-mode echocar-

diography in cardiac transplant patients with acute rejec-

tion (abstract). Eur Heart J 1988;9(suppl lk59.

13 Eichhorn P, Grimm J, Koch R, Hess OM, Carroll JD,

Krayenbuehl HP. Left ventricular relaxation in patients

with left ventricular hypertrophy secondary to aortic valve

disease. Circulation 1982;65:1395.

14 Hirota Y. A clinical study of left ventricular relaxation.

Circulation 1980;62:756.

15 Oldershaw PJ, Dawkins KD, Ward DE, Gibson DG. Dias-

tolic mechanisms of impaired exercise tolerance in aortic

valve disease. Br Heart J 1983;49:568.

16 Sutton St John MG, Tajik AJ, Gibson DG et al. Echocar-

diographic assessment of left ventricular filling and septal

posterior wall dynamics in idiopathic hypertrophic subaor-

tic stenosis. Circulation 1978;57:512.

17 Chen W, Gibson DG. Relation of isovolumic relaxation to

left ventricular wall movement in man. Br Heart J

1979;42:51.

18 Gibson DG. Brown DJ. Measurement of instantaneous

left ventricular dimension and filling in man, using

echocardiography. Br Heart J 1973;35:1141.

19 Trail1 TA. Gibson DG, Brown DJ. Study of left ventricular

wall thickness and dimension changes using echocardiog-

raphy. Br Heart J 1978;40:162.

20 Upton MT, Gibson DG, Brown DJ. Echocardiographic

assessment of abnormal left ventricular relaxation in man.

Br Heart J 1978;38:1001.

21 Park JW, Ziegler AG, Janka HU, Doering W, Mehnert H.

Left ventricular relaxation and filling pattern in diabetic

heart muscle disease: an echocardiographic study. Klin

Wochenschr 1988;66:773.

22 Bryhn M. Echocardiographic assessment of left ventricular

diastolic function in a normal population and a group of

patients with myocardial hypertrophy. Clin Cardiol

1984;7:335.

23 Btyhn M. Abnormal left ventricular filling in patients with

sustained myocardial relaxation: assessment of diastolic

parameters using radionuclide angiography and echocar-

diography. Clin Cardiol 1984;7:639.

24 Decoodt PR. Mathey DG, Swan HJC. Automated analyses

of the left ventricular diameter time curve from echocar-

diographic recordings. Comp Biomed Res 1976;9:549.

25 Bryhn M. Left ventricular diastolic function in health and

disease with special emphasis on echocardiography. Lund.

Sweden: Studentlitteratur, 1986;30.

26 Bahler RC, Vrobel TR, Martin P. The relation of heart

rate and shortening fraction to echocardiographic indexes

of left ventricular relaxation in normal subjects. J Am Coil

Cardiol 1983;2:926-933.

27 Sahn DJ, DeMaria A, Kisslo J, Weyman A. The commit-

tee on M-Mode Standardization of the American Society

of Echocardiography; recommendations regarding quanti-

tation in M-Mode echocardiography: results of a survey of

echocardiographic measurements. Circulation 1978%

1072-1083.

28 Billingham ME. Diagnosis of cardiac rejection by endomy-

ocardial biopsy. Heart Transplant 1981;1:25.

29 Brutsaert D, Rademakers FE, Stanislas U. Triple control

of relaxation: implications in cardiac disease. Circulation

1984;69:190.

30 Rousseau MF, Pouleur H, Detry JMR, Brasseur LA. Re-

lationship between changes in left ventricular inotropic

state and relaxation in normal subjects and in patients

with coronary artery disease. Circulation 1981;64:736.

31 Feigenbaum H. Echokardiographie. Deutsche Ausgabe.

herausgegeben von G. Autenrieth. Erlangen: Verlag Dr.

med. D. Straube, 1979;307.

32 Feneley M, Kearney L, Farnsworth A, Shanahan M, Chang

V. Mechanisms of the development and resolution of

paradoxical interventricular septal motion after uncompli-

cated cardiac surgery. Am Heart J 1987;114:106-114.

33 Feigenbaum H. Echokardiographie. Deutsche Ausgabe,

herausgegeben von G. Autenrieth. Erlangen: Verlag Dr.

med. D. Straube, 1979;286.

34 Valantine HA, Appleton CP, Hatle LK, Hunt SA. Stinson

EB, Popp RL. Influence of recipient atrial contraction on

left ventricular filling dynamics of the transplanted heart

assessed by Doppler echocardiography. Am J Cardiol

1987;59:1159-1163.

35 Haverich A. Scott WC, Dawkins KD, Billingham ME,

Jamieson SW. Asymmetric pattern of rejection following

orthotopic cardiac transplantation in primates. Heart

Transplantation 1984;3(4):280.

36 Billingham ME. Some recent advances in cardiac pathol-

ogy. Hum Pathol 1979;10:367.

37 Kemnitz J, Cohnert T, Schafers HJ et al. A classification

of cardiac allograft rejection. Am J Surg Pathol 1987;

11:503.