| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1995;109:1103-1115
© 1995 Mosby, Inc.
CARDIAC AND PULMONARY REPLACEMENT |
Stanford, Calif.
From the Department of Cardiothoracic Surgery, Stanford UniversityMedical Center, Stanford, Calif.
Address for reprints: William M. DeCampli, MD, PhD, Division of CardiothoracicSurgery, Children's Hospital Oakland, 747-52nd St., Oakland, CA 94609.
Abstract
The clinical status and quality of life of 40 patientswho lived or are still alive more than 10 years after transplantation at ourinstitution were reviewed with the use of our transplant database, prospectivepatient examinations, cardiac catheterization, and exercise testing. Patient-perceivedhealth status was determined with use of the Nottingham Health Profile andGeneral Well Being examinations. Factors associated with longevity were determinedby a Cox proportional hazards model. Twenty-six patients are alive and 14have died. The mean age at transplant was 32.4 ± 12 years and the currentage (or age at death) is 46.1 ± 12.8 years. Actuarial freedom fromrejection was similar to that of patients surviving less than 10 years (p = 0.8), but freedom from alltypes of infection was less (p = 0.005). Immunosuppressive drugsinclude cyclosporine (11/26 patients), azathioprine (24/26), and prednisone(26/26, mean dose 12.7 mg/day). Catheterization hemodynamic data show well-preservedgraft function at a mean follow-up of 11.7 ± 3.3 years. Graft coronaryartery disease prevalence is 51.0% ± 8%. Exercise test results areas follows: duration 8.7 ± 3.5 minutes (range 2 to 16 minutes), maximumheart rate/expected rate 77.3% ± 11% (50% to 92%), maximum systolicblood pressure 171 ± 23 mm Hg (140 to 208 mm Hg), and metabolic equivalents9.2 ± 2.3 units (5.5 to 12.9 units), or about 84% of predicted. Meanscore on the General Well Being examination was 75.3 ± 21.6 (normal).Nottingham Health Profile scores were nearly normal, except for in the 50-to 64-year-old age group in categories of mobility, pain, sleep quality, andenergy level. Causes of death were coronary artery disease in 7 of 14, infectionin 4 of 14, lymphoma in 1 of 14, and nonlymphoid cancer in 2 of 14. In theCox regression, variables most associated with survival t > 2.0, multivariate p = 0.0005) were age at transplantation(t = 3.26), preoperative duration of illness(t = 3.57),postoperative cytomegalovirus infection (t = 2.16), and ejection fraction at 12 monthsafter operation (t = -2.62). We conclude that cardiac transplantationcan provide patients with end-stage cardiac failure an acceptable generalmedical condition, functional status, and perceived quality of life well intothe second decade after operation. (J T HORAC C ARDIOVASC S URG 1995;109:1103-15)
Cardiac transplantation has proved to be a reasonable therapeutic optionfor selected patients with end-stage heart failure. This conclusion is supportedby the favorable survival rate and postoperative quality of life of patientswho have undergone cardiac transplantation, compared with these findings inthose cases managed medically. 
1 Most studies that support this conclusion, however, have had meanor median follow-up periods of only a few years and thus do not provide evidenceof the clinical status and perceived quality of life in the longer run, thatis, beyond a decade after operation. This limitation is imposed by (1) therelatively low 10-year survival of patients operated on before 1982 (approximately30% at Stanford) and (2) the existence of only a very few institutions thatperformed a significant number of transplantation operations before 1983.With more than 2000 heart transplant operations currently done per year inthe United States, and an actuarial 10-year survival rate of approximately50%, an understanding of the clinical status and perceived quality of lifemore than a decade after operation is of obvious pertinence.
In this paper we present results of a study of 40 patients who are aliveor who lived longer than 10 years after cardiac transplantation at a singleinstitution. Specifically, we tested the following hypotheses about theselong-term survivors: (1) the prevalence of graft coronary artery disease (CAD)continues to increase with time and graft CAD is a significant source of mortalityfor this group of patients, (2) graft function is well-preserved beyond 10years after transplantation, and (3) the exercise capacity of these patientsand sense of well-being are reasonably close to those of the age-matched populationat large.
Finally, we present a derivation of clinical factors that correlatewith longer-term survival in our institution's total transplantation experience.
PATIENTS AND METHODS
The preoperative clinical status and postoperative clinical course (includingregular follow-up history, physical examination, laboratory values, cardiacbiopsy results, and cardiac catheterization data) of all patients undergoingheart transplantation at Stanford since 1970 were examined with the assistanceof a database updated to January 1, 1992. The subset of patients who survivedmore than 10 years after their first heart transplant operation was examinedin detail. Of the 661 patients in the database, 40 patients fell into thissubset: 26 of these patients were alive and 14 had died. Seven of the 40 patientsunderwent cardiac retransplantation. All but two of these received retransplantationmore than 10 years after the original operation. One patient had undergoneretransplantation twice. The patients and their dates of transplantation arelisted in Table I.
|
So that functional status could be assessed, exercise tests were doneon available patients. The actual exercise protocol varied for each patient;however, a "maximal" test was completed and maximum heart rate, blood pressure,duration, and exercise capacity (expressed in metabolic equivalents [METS]standardized for each exercise protocol) were determined. Results were comparedwith those of an age-matched sample of the normal population and with thoseof an age-matched cohort of patients who underwent transplantation an averageof 7 months before evaluation.
So that we could evaluate patient-perceived health status or "senseof well-being," each patient in the subset completed both part I of the NottinghamHealth Profile (NHP) and the General Well Being (GWB) examinations. The NHPexamination is purported to evaluate "perceived distress." 2,3 The higherthe score (0 to 100 for each of six sections), the more the distress. Weightingsderived from the application by McKenna, Hunt, and McEwen 4 of Thurstone's method of paired comparisons to theBritish population were applied to each question in the NHP survey. The GWBexamination is best described as reflecting "expressions of psychiatric statesof depression and anxiety or the absence thereof," according to Fazio. 5 Scores from both tests were comparedwith those of published age and gender-matched norms. 6 Correlations were calculated among the NHP and GWBscores, treadmill performance, employment and marital status, and number ofcurrent medical problems.
Because we were interested in factors that might contribute to (or atleast be correlated with) the longevity of this subset of patients, two comparisongroups were formed from the set of all patients who underwent transplantationat Stanford: (1) patients who survived 0.5 to 5.0 years after transplantationand (2) patients who survived 5 to 10 years after transplantation. The ageat the time of transplantation; actuarial rates of freedom from acute rejectionand from viral, bacterial, protozoal, and fungal infection; and cause of deathwere determined for each group and compared.
Finally, a formal multivariate analysis was done to further clarifythe possible factors. Twenty-one variables were cited as possibly predictiveof longevity. Variables evaluated at the time of transplantation includedduration of preoperative illness, age at the time of transplantation, race,gender, hospital length of stay, number of human leukocyte antigen mismatches,serum creatinine level at the time of discharge, diastolic blood pressureat the time of discharge, donor age, donor race, donor gender, donor cytomegalovirus(CMV) status, and recipient CMV status. Variables evaluated at 1 year afteroperation included blood urea nitrogen, serum creatinine, diastolic bloodpressure, left ventricular ejection fraction, degree of CAD (quantitated bythe number of vessels with significant stenoses), number of acute rejectionepisodes, number of episodes of infection, and incidences of CMV infection(as determined by the presence of inclusion bodies in tissue, positive culture,or greater than fourfold rise in anti-CMV immunoglobulin G serologic titer).Univariate analysis of variance was performed for each variable from the aforementionedthree groups to obtain a qualitative estimate of significance. The nine mostsignificant variables were determined. The set of all patients who underwenttransplantation at this institution was then chosen and a subset determinedthat consisted of patients each meeting all of the following three criteria:(1) survival at least 1 year after operation, (2) not having retransplantation,and (3) having no missing data among the nine variables. By including onlypatients who survived at least 1 year, we eliminated operative and early postoperativecomplications and early refractory rejection and infection as causes of mortality.The final subset, consisting of 122 cases, was used in the multivariate analysis(see following section).
Results are expressed as mean plus or minus the standard deviation,unless otherwise specified. Comparison of means was made by one-way analysisof variance or Student's t test. The Fisherexact test was used for nominal data in the analysis of the subset of patientsreceiving cyclosporine (n = 8), and the Mann-WhitneyU test was used to compare variables between patients with and without evidenceof CAD. Actuarial calculations were made with use of the product limit estimatorwith Greenwood's standard error. 7 Comparison of actuarial curves was made by the Cox-Mantel test. Correlationswere examined by determining Pearson coefficients or Spearman rank correlationcoefficients. A value of p < 0.05 was consideredsignificant. The multivariate analysis was done by the stepwise proportionalhazards method of Cox. 8 Atest of parallelism was made to ensure that the final covariates induced proportional(parallel) shifts of the arbitrary hazard function. A 
2 statisticwas used to test the hypothesis that the gradient of the log-likelihood functionwas zero at its initial value, and p <0.05 was considered significant. The relative risk of death caused by an increaseby m units in a covariate was calculated asexp (mb), where bis the estimate of the normalized coefficient. Ninety-five percent confidencelimits of the relative risk are given.
Henceforth the expression most recentrefers to the most recently determined value of avariable in living patientsor the last-determined value before death in deceased patients.
RESULTS
Dates of transplantation and retransplantation for the group of patientssurviving longer than 10 years are shown in 
Table I.The majority (32/40) underwent transplantation before the introduction ofcyclosporine into our program in December 1980. Basic demographic data arelisted in Table II. In comparison with patients in the short survival(0.5 to 5 years) group, the patients in the group with more than 10-year survivalreceived transplantation at a younger mean age. In the former group, only15 of 119 patients were younger than 30 years old at operation versus 18 of40 for the long-term survivors (p < 0.001).
|
|
|
Most recent laboratory values for the patients are shown in TableIII. Of the eight patients with serum creatinine valueshigher than 1.8 mg/dl, seven were taking cyclosporine and four of these sevenhad undergone retransplantation. These values are significantly higher thanone would expect if the eight patients were a random subset of the group (Fisher'sexact test, p < 0.01). The mean total cholesteroland triglyceride values are influenced by one patient with isolated hypertriglyceridemia(most recent measurement 468 mg/dl) and by two patients with combined disorders(cholesterol 445 and 293 mg/dl, triglycerides 273 and 298 mg/dl, respectively).For five patients evaluated in the retransplantation subgroup, values significantlyhigher than those in the first-time transplantation subgroup included bloodurea nitrogen (36.0 ± 7.9 mg/dl, p= 0.01), creatinine (2.0 ± 0.27 mg/dl, p= 0.002), and total cholesterol (257.2 ± 33.7 mg/dl, p = 0.02).
|
TableIV. A remarkable feature is the relatively normal hemodynamicdata, indicating, on average, well-preserved graft function at this long follow-upinterval. Aortic diastolic pressure and left ventricular end-diastolic pressurewere mildly elevated. The prevalence of CAD was 51.0% ± 8.0% (standarderror of the estimate) at a mean of 11.7 ± 3.3 years of angiographicor autopsy follow-up. For each patient, we determined the earliest postoperativetime in which angiographic evidence of graft CAD was detected: this mean valuewas 9.3 ± 4.1 years (n = 40). The actuarialfreedom from CAD for the study group is shown in Fig.3 together with that for all other patients who underwent transplantationat Stanford. These two curves are not statisticallysignificantly different. Values for number of acute rejection episodes, aorticdiastolic pressure, serum cholesterol, current age, current weight, left ventricularend-diastolic pressure, and NHP score were not significantly different betweenthe subgroups of patients with and without evidence of CAD.
|
|
Table V. Patients with heart failure or poor overall medicalcondition tended to refuse the test and thus the results are biased on thehigh side. One test was stopped because of electrocardiographic evidence ofischemia and the remainder were patient-terminated because of fatigue. Therewas a significant correlation between METS performed and cardiac index onmost recent cardiac catheterization (Spearman coefficient 0.81, p = 0.02).
|
The actuarial survival curve for the group is shown in Fig4, together with the analogous curve for all patients who received transplantationat this institution (evaluated on March 1, 1994). The causes of death for the 14 patientswho died are displayed in Fig. 5, along with those ofpatients in the comparison group who survived 0.5 to 5 years. These data include the three patientsundergoing retransplantation who died. The predominant cause of death wasCAD in the long-term survivors, with no deaths attributable to complicationsof acute graft rejection. In the short-term survivors, rejection and infectionplay a dominant role in mortality.
|
|
Table VI. The four variables found to be independently correlatedwith death after transplantation were age at transplantation, preoperativeduration of illness, evidence of postoperative CMV infection, and ejectionfraction at 12 months after operation. The relative risk of death associated with an increaseof 5 years in the duration of illness was 1.53 (1.21, 1.93). An increase of5 years in the age at transplantation carried a relative risk of 1.28 (1.10,1.48). A decrease in the ejection fraction at 12 months of 0.1 resulted ina relative risk of death of 1.51 (1.10, 2.06). History of postoperative CMVinfection approximately doubled the risk of death (1.07, 3.96). As an independenttest of the risk associated with CMV infection, life-table calculations usingthe product limit estimator on the set of all patients undergoing transplantationat this institution were stratified by presence or absence of evidence ofpostoperative CMV infection. The probability of death within 6 years afteroperation was 26.7% without CMV infection and 46.0% with CMV infection, arelative risk of 1.72 (1.20, 2.24).
|
Although justification of cardiac transplantation as an effective treatmentfor end-stage heart failure is provided by prolonged (>2 year) survival andimproved functional status, the argument in its favor is greatly strengthenedby demonstration of the possibility of longer-term survival with sustainedacceptable functional status. This demonstration was precisely how the currentimportance of renal transplantation was established 15 years ago. It is appropriate,therefore, to discuss our results in the context of this paradigm.
General medical condition
Aside from rejection and graft CAD, most of the aforementioned medicalproblems seen in our transplant recipients are probably complications of immunosuppressivetherapy. Divakar and colleagues 9 found a similar variety and incidence of problems in a group of 36patients surviving greater than 10 years after renal transplantation. Thespecific role that corticosteroid usage plays in these problems has led manyprograms to use corticosteroid-free maintenance immunosuppression. Initialresults of this regimen are encouraging, 10,11 although the regimen is notfeasible in many patients. We found that the incidence of chronic renal insufficiencywas highly correlated to the use of cyclosporine in our group. This problemstill remains, although trials using "low-dose" cyclosporine 12 and cyclosporine analogs such as FK 506 may holdsome promise. The incidence of infection in our long-term group was remarkablylow relative to that in the two comparison groups, an observation consistentwith the fact that infection is a major cause of early death with immunosuppressivetherapy. The fact that the curves of actuarial freedom from acute rejectionin this group and both comparison groups were all nearly identical means,however, that other factors influenced the group's susceptibility to earlyfatal infection, such as differences in severity or frequency of rejectionor exposure to pathogens, degree of augmentation of immunosuppressive drugs,or individual differences in the degree to which given doses of immunosuppressivedrugs influence the host response to pathogens.
Graft function
The present work demonstrates, for the first time, preserved graft functionwell into the second decade after transplantation. Von Scheidt and associates 13 found no evidence of systolicdysfunction or dilated or restrictive cardiomyopathy in 71 patients followedup to 7 years after operation. Tischler and associates 14 demonstrated mild increases in left ventricularend-systolic and end-diastolic volumes, but no significant change in leftventricular mass or ejection fraction in 22 patients doing well and followedup a mean of 4 years after operation. Frist and colleagues, 15 in a study of 174 patients with a mean age of 36years and follow-up period of 5 years, found no deterioration in graft hemodynamics.Mean cardiac index and left ventricular end-diastolic pressure were almostidentical to those in the present group.
Exercise tolerance
Although the exercise tests described in this paper were not standardizedto a particular protocol, certain points can be made. First, maximum heartrate and systolic blood pressure were lower than those of a closely age-matchedsample of the normal population (p < 0.001), 16 but were similar to those of anage-matched group of transplant recipients evaluated a mean of 7 months afteroperation in a study by Kavanagh and associates. 17 These results may be explained by the failure ofthe graft to regain sympathetic innervation a mean of 12 years after implantation.Second, when each patient's estimated maximal oxygen consumption (expressedin METS) is compared with the mean value from an age-matched group of normalsubjects not having transplantation, the group performed an average of 1.19± 0.60 (SEM*) METS less (p = 0.08) than normal subjects as determined by Bruce, Kusumi,and Hosmer 18 and 2.55 ±0.65 (SEM) METS less (p = 0.002) than normalsubjects as published more recently by Morris and colleagues. 19 These results correspond to an exercise capacityof 88% ± 6% and 79% ± 6% of normal, respectively. Yet the group'smean performance was substantially better than the average of 6.2 ±1.3 METS performed in the group of transplant recipients at 7 months afteroperation in the report of Kavanagh and associates. 17 A possible interpretation of these data is thatexercise capacity improves with duration after operation, which is a hypothesisthat could be better tested by a longitudinal study. Alternatively, the datamay indicate a time-related bias in patient selection. The cause of the remainingdecrement in exercise capacity at 12 years after operation is uncertain, 20 but it may be compensated for,as Kavanagh and associates 17demonstrated, by an exercise training program.
Graft CAD
The most remarkable finding from the angiographic data is that thisgroup of long-term survivors does not escape graft CAD. In fact, the actuarialincidence during the first 10 years after operation is similar to that ofthe transplant population as a whole. Furthermore, the data indicate thatby 20 years after operation essentially all patients can be expected to havegraft CAD identifiable by conventional angiographic techniques. The comparisonbetween our group and the overall transplant population is made with somereservation; that is, because the overall transplant population is right-censoredby death in the first 10 years, its actuarial prevalence of CAD will be underestimatedrelative to the estimate for our group. Studies that have attempted to determinerisk factors for development of graft CAD give inconsistent results, 21-29 and the reported prevalence of CAD varies widely. 27,30-32 Indeed, we were unable to associate any variableswith the presence of CAD in our small group of patients.
Psychosocial evaluation and perceived quality of life
In previous, shorter-term studies in this category, the principal findingswere (1) overall sense of well-being improves after transplantation relativeto the preoperative state, 33-35 (2) on the average, thereis no deteriorationof quality of life with time, 33,35,36 (3) perceived quality does not correlate significantly withmedical or demographic data, 33 (4) a substantial minority of patients have problems that decreasequality of life, especially weakness, sleep disturbances, and sexual dysfunction(impotence), 36-40 (5) the preoperative perceived quality of lifeaffects the postoperative quality of life and perhaps survival, and (6) about50% regain employment after operation.
Several of these findings seem to hold for the present group of patientsevaluated more than 10 years after transplantation. The trends in the NHPscores are in accord with those of a large study by Rosenblum and associates, 38 who found that a group of transplantrecipients with a median age of 53 years had Sickness Impact Profile scores"similar to those of cardiac arrest survivors and postmyocardial infarct patients."On the other hand, Aravot and colleagues 41 found much lower NHP scores at a mean of 22 months of follow-up ina group of patients undergoing transplantation after age 60 years. It couldbe that a combination of older age and longer postoperative duration correlatewith an increasing perception of distress. The percentage of patients employedis similar to that in other reports, indicating that there does not appearto be a decline of employment with duration after operation. As in prior reports, 6 we could find no correlation betweenmedical data and perceived quality of life. In the future, longitudinal studies(such as those done by Jones 33 and Caine 35 and theirassociates) will permit a better assessment of the long-term, time-dependenttrends in these psychosocial and quality-of-life parameters.
Multivariate analysis of survival
Age at the time of transplantation has been cited previously as a correlatewith survival duration. 1This result should be interpreted with caution, however, because patientswho received transplant earlier in our experience were younger at the timeof transplantation than were patients who received transplant later in ourexperience. This bias means that relatively more right-censoring exists inthe survival function of the older patients. A second problem with the interpretationof the age factor is that the instantaneous death rate of patients undergoingtransplantation at older ages (>50 years old) is significantly influencedby that of the age-matched nontransplanted population. This means that deathbeyond this age may well not be related to the sequelae of heart transplantation.
The duration of illness before transplantation was an intriguing correlateof survival. Evans and associates 42 found that higher levels of pretransplantation functional impairment,as measured by the Karnofsky index, correlated strongly with poorer posttransplantationsurvival. O'Brien, Buxton, and Ferguson 6 found that patients indicating fewer quality-of-life restrictionsbefore transplantation had a better likelihood of long-term survival. Thedegree to which preoperative duration of illness correlates with these twofactors is unclear. However, a relationship is suggested by the fact thatcongestive heart failure is usually progressively debilitating. It could bethat subtle aspects of debilitation resulting from cardiac failure are notreversed with cardiac transplantation or are perhaps worsened by the adverseeffects of immunosuppressive therapy.
In a recent study by Elkins and associates, 43 primary or recurrent CMV infection occurred in 64%of heart transplant recipients a mean of 164 and median of 84 days after operation. Some of the decrease in survival as a result of CMV infection can be accountedfor by various manifestations of the disease or syndrome. The correlationof a history of CMV infection with survival and with CAD incidence was firstnoted by Grattan and colleagues 44 in 1989. CMV can regulate major histocompatibility complex classI expression on smooth muscle cells, 45 and CMV infection is associated with progressive intimal thickeningof intramyocardial arterioles in cardiac grafts. 46 It is thus likely that the late effects of CMV infectionon survival are due to its contribution to graft CAD. This notion is supportedby our multivariate analysis, that is, partly because graft CAD is not usuallydetectable in the first 12 to 24 months by conventional angiographic analysis,CMV infection, and not graft CAD per se, was an independent risk factor formortality.
The ejection fraction at 12 months was the fourth significant correlatewith survival in our analysis. This result is not surprising. Myocardialdysfunction can result from a variety of causes, such as severe recurrentrejection, early graft CAD, myocarditis (often viral or drug-induced), cardiomyopathy,or "nonspecific" graft failure. In our analysis, ejection fraction at 12 monthscorrelated somewhat with detection of CAD at 12 months (Pearson coefficient= -0.35), but not with diastolic blood pressure at 12 months, the total numberof episodes of acute rejection, or the incidence of all viral infections.
With respect to the aforementioned multivariate analysis, it is importantto note the limitations of the proportional hazards method. Although it isrelatively easy to apply, it assumes that all covariates will have the sameshape of hazard function. A more versatile approach would determine each hazardfunction, thus perhaps allowing some understanding of the biologic natureof the dependence of the covariates in the model.
The vast majority of this group of patients underwent operation in anera when many of the improved methods of patient management now in use wereunavailable, such as cyclosporine, "steroid-free" maintenance immunosuppression,specific anti-T cell induction immunosuppression, routine early antibioticprophylaxis against CMV (ganciclovir) and Pneumocystiscarinii (trimethoprim/sulfamethoxazole), third-generation antibiotics,early use of prophylactic calcium channel blockade (diltiazem) for reductionof graft CAD progression, 47improved donor selection, and other empiric features gained from accelerationof experience in cardiac transplantation worldwide after about 1983. Thusthe conclusions we have drawn from our study are likely to be unique to thisgroup.
Conclusions
Results of this study of long-term survivors of heart transplantationsupport the following conclusions: (1) graft function is well-preserved intothe second decade after transplantation, (2) exercise capacity is 80% to 90%of that of the age-matched general population, (3) patient-perceived healthand well-being (by NHP and GWB examinations) is close to that of the age-and sex-matched general population, (4) the cumulative incidence of graftCAD is similar to that of the cardiac transplant population as a whole, continuesto develop after 10 years, and is a significant source of mortality, and (5)long-term survival is correlated with four factors that can be evaluated by1 year after transplantation; younger age at time of transplantation, shorterpreoperative duration of illness, absence of postoperative CMV infection,and larger left ventricular ejection fraction at 1 year after operation.
Cardiac transplantation can provide patients with end-stage cardiacfailure an acceptable general medical condition, functional status, and perceivedquality of life well into the second decade after operation
Appendix: DISCUSSION
Dr. Alfredo Trento (Malibu, Calif.).
The importance of this review is not in the comparison between different
groups of patients undergoing transplantation. These were groups of patients
who received transplantation in different periods of time when there were
different immunosuppressive regimens and in which there were very different
understandings about the side effects of immunosuppression therapy and about
how to prevent the side effects. The conclusions are extremely important and
are those that we transplant surgeons have been expecting.
The conclusions are that the quality of the life for the surviving patients is similar to that of an age-matched population and that the main cause of late death beyond 10 years after transplantation is CAD. CAD is progressive to the point that if we wait long enough every transplant recipient would be affected. The other conclusion is that rejection and CAD together are the main cause of death in the patients who died within 10 years of transplantation. This brings me to my first question. In the manuscript the authors reported that only 15% of the patients are free of rejection 6 months after transplantation. I reviewed our experience and found that at 1 year after transplantation 60% of the patients are free of rejection and 60% of them have also been weaned from steroids. Can you clarify this high prevalence of rejection, and do you think that played a role in the long-term survival and in the incidence of CAD?
Dr. DeCampli.
I think the important factor here is that these patients received transplantation
in a unique interval of time in which we were on a very steep learning curve
to understanding how to best manage transplant cases preoperatively, immediately
postoperatively, and in the long term. The patients underwent transplantation
between 1970 and 1982. Only a few of them ever received cyclosporine. It was
probably the absence of cyclosporine, and not so much the absence of anti-CD4
induction therapy, that accounted for the high incidence of rejection in the
first couple of months after operation. I qualify that by saying that even
with the addition of cyclosporine in the modern era and with the addition
of OKT3 routinely, Stanford's incidence of rejection in the first 6 months
is still 75% to 80%. That number does differ from yours and I cannot, offhand,
explain that difference.
Dr. Trento.
The manuscript points out the effect of CMV infection on survival: history
of postoperative CMV infection doubles the risk of death at 6 years after
transplantation. It is clear now that CMV infection seems to trigger some
sort of generalized immunologic reactivation that may trigger a form of acute
or chronic rejection that we do not know how to diagnose or to treat. This
may play a role in the occurrence of CAD. For this reason, we have started
at our institution an aggressive and also costly type of prophylaxis against
CMV infection, which involves the administration of hyperimmune gamma globulin
weekly for the first 4 weeks and then intravenous administration of ganciclovir
for 2 weeks followed by oral acyclovir for 3 months. With this regimen we
have decreased the incidence of postoperative CMV infection from a range of
about 35% to 70% as reported in the literature and from a range of about 50%
reported in this paper to 4% in patients who were CMV positive before transplantation
and to about 26% in the difficult group of patients who were CMV negative
before transplantation and received a CMV-positive organ. Would you comment
on this aspect and also basically on the approach that Stanford is taking
at the present for CMV prophylaxis?
Dr. DeCampli.
Stanford's approach currently is as follows. If I designate D as donor
and R as recipient and D-negative R-negative means that neither was CMV positive,
then we have no prophylaxis for D-negative R-negative situations. In the D-negative
R-positive and D-positive R-positive situations, we give ganciclovir for 2
weeks intravenously and for 2 weeks orally for a total of 4 weeks of therapy.
With this regimen we have achieved a significant reduction, as compared with
the results of these long-term survivors, in CMV infection in the first 6
months. Thus the incidence is now approximately 15% rather than 50%. In the
so-called CMV mismatch group, that is, D-positive R-negative, which is the
most difficult group, we have found no benefit with 4 weeks of ganciclovir
therapy. Stanford has empirically extended its coverage from 4 to 6 weeks.
Stanford has found no particular benefit to the use of CMV antiglobulin, in
particular in the D-positive R-negative group. We have had no early deaths
from CMV infection in the past 5 years. However, we cannot yet make a statement
on the effect of any of these prophylaxis regimens on the later incidence
of graft CAD. I am not sure any institution can do that yet.
Dr. Trento.
It is my feeling that any immunosuppression based on cyclosporine or
cyclosporine analogs has reached a plateau and basically we have gotten out
of cyclosporine everything that cyclosporine can give. Further improvement
in results will probably be only minor unless a qualitative change in the
immunosuppressive therapy is made. Would you like to comment on this last
point?
Dr. DeCampli.
If you ask how we can best improve the survival of patients undergoing
transplantation and if you make the assumption that graft CAD is caused by
a chronic rejection phenomenon, then the answer is more selective immunosuppression
leading ultimately to what is now popularly known as "organ-specific clonal
anergy." One possible way of achieving this is the use of total lymphoid irradiation
for tolerance induction. The manner in which stem cells regenerate after total
lymphoid irradiation is in a peculiar order that, in fact, favors acceptance
of a transplanted organ; that is to say, suppressor cells seem to recover
first after total lymphoid irradiation, blunting the effect of the particular
lymphoid subset that acts against the implanted organ. Total lymphoid irradiation,
in conjunction with the use of a new anti-CD4 antibody that actually depletes
CD4 cells (rather than just altering their surface membrane structure), is
currently being investigated as a combination that could give us more tolerance
induction and organ-specific clonal anergy. Dr. Bruce Reitz is working on
this problem in the laboratory at Stanford currently with primates.
Footnotes
Read at the Twentieth Annual Meeting of The Western
Thoracic Surgical Association, Olympic Valley, Calif., June 22-25, 1994. 
References
This article has been cited by other articles:
|
|
 |
|
  T. Hussain, M. Burch, M. J. Fenton, P. M. Whitmore, P. Rees, M. Elliott, and P. Aurora Positive Pretransplantation Cytomegalovirus Serology Is a Risk Factor for Cardiac Allograft Vasculopathy in Children Circulation, April 3, 2007; 115(13): 1798 - 1805. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. John, H. A. Rajasinghe, S. Itescu, S. Suratwalla, K. Lietz, A. D. Weinberg, A. Kocher, D. M. Mancini, R. E. Drusin, M. C. Oz, et al. Factors affecting long-term survival (>10 years) after cardiac transplantation in the cyclosporine era J. Am. Coll. Cardiol., January 1, 2001; 37(1): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Hetzer, W. Albert, M. Hummel, M. Pasic, M. Loebe, H. Warnecke, A. Haverich, and H. G. Borst Status of Patients Presently Living 9 to 13 Years After Orthotopic Heart Transplantation Ann. Thorac. Surg., December 1, 1997; 64(6): 1661 - 1668. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
