(Circulation. 1995;91:1029-1035.)
© 1995 American Heart Association, Inc.
Articles |
From the Cardiovascular Clinical Pharmacology Laboratory and Center for Women's Health, Department of Medicine (M.E.W., E.-G.V.G., R.R.S.), and the Division of Cardiothoracic Surgery, Department of Surgery (E.A.R., M.L.B.), College of Physicians and Surgeons, Columbia University, New York, NY, and Health Care Consultation Center (M.L.B.), USC School of Medicine, Los Angeles, Calif.
| Abstract |
|---|
|
|
|---|
Methods and Results The study population consisted of 379
patients (75 women, 304 men)
18 years of age who survived for
48
hours after undergoing orthotopic cardiac transplantation between March
1985 and March 1992. The following were analyzed: incidence of death
and treated rejection episodes, donor and recipient cytomegalovirus
(CMV) matches, use of OKT3 induction therapy, and donor and recipient
HLA mismatches. Women 49±12 years old and men 47±12 years old
were
characterized by differences in race and diagnosis. Women were more
likely to be nonwhite (P<.01) and have idiopathic
cardiomyopathy than were men (P<.01). A trend toward an
increase in first-year rejection frequency was seen in women compared
with men (P=.08). Overall actuarial survival was
significantly reduced in women after transplantation
(P<.05). At 36 months, female actuarial survival was
64±7% versus 76±3% for men (P<.05). The majority of
patients in this study did not receive CMV prophylaxis. Univariate
analysis revealed that only CMV(+) donor status and the use of OKT3
induction therapy affected survival in women. Multivariate analysis
revealed a marked reduction in survival in female recipients of CMV(+)
donors given OKT3 induction therapy. At 36 months, only 25% of women
were still alive compared with 86% of women with neither risk factor
(P<.001). Even without OKT3 induction there was markedly
reduced survival in women with mismatched CMV status, ie, CMV(-)
recipients of CMV(+) donors; 17% survival after 36 months versus 86%
in women who were CMV(+) recipients (P<.05). Although at
this institution during the study time period, CMV prophylaxis was not
routinely employed and OKT3 induction was selectively used in
higher-risk patients, conclusions regarding differences in outcome that
are sex dependent are valid.
Conclusions (1) Women are at risk for reduced actuarial survival up to 3 years after cardiac transplantation. (2) Univariate analysis shows that women are selectively at risk for death when receiving hearts from CMV(+) donors and after receiving OKT3 induction therapy. (3) Multivariate analysis reveals that women are at even greater risk for death when receiving hearts from CMV(+) donors in conjunction with OKT3 induction therapy. (4) In the absence of OKT3 use, the greatest risk of death occurs in CMV(-) women transplanted with CMV(+) donor hearts. (5) When female to male survival curves are compared, factors that influenced survival in women did not appear to be problematic in men.
Key Words: risk factors transplantation clinical trials rejection
| Introduction |
|---|
|
|
|---|
Due to the relatively small number of women undergoing cardiac transplantation, few studies have clearly examined the effect of gender on cardiac transplantation. Among reported studies, there are conflicting data on rejection and survival. Two small studies, of 21 and 27 women, respectively, concluded that women have a higher incidence of rejection after cardiac transplantation.3 4 In a larger population of 70 women, Zerbe and coinvestigators5 found that rejection in women is comparable to that found in men. Another report from Sharples and associates6 found that women are at increased risk of death from rejection. Although the International Heart Transplant Registry reported decreased actuarial survival in women compared with men in the 1988 report, it was not decreased in the 1989 report.1 7
The present study was undertaken to evaluate the risk factors for mortality and morbidity in women after cardiac transplantation. It is a retrospective analysis of survival and rejection after cardiac transplantation in 75 women and 304 men. The study seeks to examine factors that might explain sex-related differences after cardiac transplantation.
| Methods |
|---|
|
|
|---|
48 hours and were >18 years old. Only patients
surviving 48 hours were analyzed to exclude surgical mortality. Data
were obtained by review of each patient's chart, as well as review of
autopsy and pathology reports to verify rejection, survival, and cause
of death. All patients were maintained on triple immunosuppressive
therapy consisting of cyclosporine, azathioprine, and prednisone.
Beginning in 1988, OKT3 induction, 5 mg/d for 7 to 14 days, was used
for higher-risk patients requiring preoperative intravenous inotropic
drugs or having an elevated serum creatinine (>1.5 to 1.8
mg/dL).8 Since 1988, 21 of 67 women and 103 of 294 men
were treated with OKT3 induction. Death and rejection were the only end
points for the study.
Risk Factors
Potential risk factors for mortality and
rejection investigated
included sex, age, race, cardiac diagnosis, cardiac classification
before treatment by New York Heart Association (NYHA) criteria and
cardiac classification by United Network for Organ Sharing (UNOS)
criteria, cytomegalovirus (CMV) serology of the donor, CMV serology of
the recipient, use of OKT3 induction therapy, ABO blood types, HLA
types, comorbid disease, prior cardiac surgery, and parity.
Rejection
Rejection was diagnosed and graded by
endomyocardial biopsy in
combination with clinical criteria. All acute rejection episodes
[International Society for Heart and Lung Transplantation (ISHLT) IIIa
or greater] were treated and patients with infiltrates (ISHLT Ib or
II) were treated on the basis of their clinical condition and the
hemodynamic data at the time of catheterization. For the purposes of
this study, rejection episodes are defined as those episodes that
required intervention and were treated with at least augmented
corticosteroids.
Cytomegalovirus Status
CMV serology of donor and recipient
was screened before
transplantation to determine the presence or absence of anti-CMV
antibodies. CMV prophylaxis was begun in May 1989 for all newly
transplanted patients receiving organs from CMV(+) donors. Prophylaxis
consisted of Sandoglobulin (Sandoz) or Gammagard (Baxter) globulin
preparations (eight doses over 12 weeks) and oral acyclovir.
Ganciclovir was used only to treat documented CMV infections and was
not part of the prophylactic regimen.
HLA Typing
HLA-A, HLA-B, and HLA-DR type was ascertained
before
transplantation in the recipient and after transplantation in the donor
in the majority of patients. The total number of HLA-A, -B, and -DR
mismatches found between donor and recipient were retrospectively
determined.
Statistical Analysis
Data are reported as means and standard
deviations for
continuous variables and as percent prevalence for discrete variables.
Group differences in continuous variables were assessed by a
t test (comparison between two groups) or by ANOVA
(comparisons among three or more groups). Differences in the prevalence
of discrete variables were analyzed by Fisher's exact test or by
2. Tests for group differences of nonnormally
distributed continuous variables, ie, rejection frequency, were
performed with the Kruskal-Wallis nonparametric
procedure.9 Kaplan-Meier estimates of survival were
computed, and univariate and multivariate tests of the significance of
group differences were performed with the log-rank
test.10
| Results |
|---|
|
|
|---|
|
Overall Survival and Rejection
Overall actuarial survival was
significantly reduced in women
after transplantation (P<.05). In this study, 26 of 75
women (35%) and 77 of 304 men (25%) died. At 6 months after
transplantation, female actuarial survival was 75±5% versus
84±2%
for men. At 36 months after transplantation, female survival was
64±7% versus 76±3% for men (P<.05) (Fig
1
). Eighteen female deaths occurred within a 6-month
period and 20 within the first year. Primary cause of death is listed
in Table 2
. No significant differences between primary
cause of death between women and men were seen. In women and in men the
number of HLA-A, HLA-B, and HLA-DR mismatches did not correlate with
increased mortality. There was no significant difference in mortality
between women and men with four to six HLA mismatches. However, women
with four to six HLA mismatches had significantly increased first-year
rejection frequency (0.26±0.41 episodes per patient month) compared
with men with four to six HLA mismatches (0.15±0.27 episodes per
patient month) (P<.05). In men, no significant difference
in first-year rejection frequency was seen in those with zero to three
HLA mismatches compared with men with four to six HLA mismatches.
|
|
Univariate Analysis of Factors Affecting Mortality
Univariate
analysis of the following risk factors affecting
survival in women and men was performed: age, diagnosis, race, NYHA
classification, UNOS classification, CMV serologies of donor and
recipient, use of OKT3 induction therapy, blood type, HLA mismatch,
and, in women, parity. Only CMV donor status and the use of OKT3
induction therapy were found to affect survival in women.
Therefore, further multivariate analy- sis was performed to determine
the interaction of CMV donor status and the use of OKT3 induction
therapy in women and men.
Multivariate Analysis of CMV, OKT3, and Sex on Survival
A
marked reduction in survival was seen in female recipients of
CMV(+) donors given OKT3 induction (n=11; Fig 2
).
At 36
months, only 25% of female recipients of CMV(+) donors given OKT3
induction were still alive versus 86% in women with neither risk
factor (n=15) (P<.001). A reduction in survival was also
noted in female recipients of CMV(+) donors not treated with OKT3
(n=24), 52% compared with 86% in those with neither risk factor;
however, this was of borderline significance (P=.06).
Survival in female recipients of CMV(-) donors who received OKT3
(n=10) was not different from those with neither risk factor, 80%
versus 86% (P=NS).
|
Survival was slightly reduced in
male recipients of CMV(+) donors given
OKT3 induction (n=54), 65% compared with 85% in men with neither risk
factor (n=76); however, this was of borderline significance
(P=.09; Fig 3
). No difference in survival was
seen in male recipients of CMV(+) donors not given OKT3 (n=65) nor
male
recipients of CMV(-) donors who were given OKT3 (n=49) compared
with
male recipients of CMV(-) donors not treated with OKT3.
|
Effect of CMV Mismatching
In women not treated with OKT3
induction, additional subgroup
analysis of CMV mismatching was performed. Markedly reduced
survival was observed in female CMV(-) recipients of CMV(+) donors
(n=8), 17% after 36 months compared with 86% survival in female
CMV(+) recipients of CMV(+) donors (n=16)
(P<.05). This
analysis was precluded in those patients given OKT3 due to the
limited number of patients in each of the subgroups.
In men not treated with OKT3 induction, there was no significant difference in survival between CMV(-) recipients of CMV(+) donors (n=18), 77% at 36 months, and CMV(+) recipients of CMV(+) donors (n=46), 76% at 36 months.
Multivariate Analysis of CMV, OKT3, and Sex on Rejection
Increased first-year rejection frequency was observed only in
women transplanted with CMV(+) donors irrespective of OKT3 use (Table
3
).
|
| Discussion |
|---|
|
|
|---|
CMV as a Risk Factor for Adverse Outcome
Clinically it has
been recognized that mismatched CMV status is
problematic for cardiac transplantation.14 15
Theoretically, with the advent of improved immunosuppression,
prophylaxis, and treatment with antiviral agents, this problem is
manageable. Reports on therapy, however, frequently discuss outcome in
morbidity terms but do not report effects on
survival.16 17 18 19 Even in
renal transplantation, trials
conducted with immunosuppressive and antiviral agents limit end points
to morbidity.20 21 Conducted within a population
treated
with triple immunosuppressive therapy, the present study strongly
suggests that mismatched CMV status in women places them at a
significantly increased risk for a fatal outcome. Is mismatched CMV
status for women (1) primarily a marker for rejection and decreased
survival or is it (2) causally related to CMV infection and a
subsequent adverse outcome? The association of CMV infection with a
variety of infections, increased rejection, increased incidence of
cardiac allograft atherosclerosis, and death after cardiac
transplantation has been
recognized.22 23 24 During
transplantation both the allograft and blood products can transmit
exogenous virus from infected donors to recipients. The mechanism(s)
responsible for CMV infection at transplantation include (1)
transmission by donor organ or via transfusion of CMV(+) blood to a
CMV(-) recipient, (2) reactivation of latent CMV infection in an
immunocompromised host, and (3) transmission of a different strain of
CMV to an already seropositive CMV recipient.18 Matching
donor and recipient with respect to CMV serology should minimize
primary infection with CMV25 26 and should therefore
improve survival. In 1988, the group at Papworth26 27
described CMV infection in a series of 33 heart-lung transplantation
recipients. Of 8 patients who received grafts from CMV(+) donors, 5
developed severe CMV pneumonitis, and an additional patient developed
gastrointestinal CMV. Death occurred in 3 of 5 patients with CMV
pneumonitis, leading to the adoption of a policy of obligatory CMV
donor-recipient matching. After implementation of the policy, morbidity
and mortality from CMV infection declined. To date, review of sex in
relation to CMV status and survival in heart transplant recipients has
not been performed. The results of the present study show an
increase in rejection episodes and an alarming and significant
reduction in survival in female recipients of CMV(+) donors up to 36
months after heart transplantation. Since this was a retrospective
study, we could neither evaluate CMV serology posttransplantation nor
perform molecular studies to determine if CMV strains found in the
donor were also found in the recipient. The possibility that primary
CMV infection contributed to the poorer outcome of CMV(-) female
recipients cannot be excluded. Primary CMV infection, particularly when
transmitted by the donated organ, is associated with significantly
higher morbidity than infection caused by secondary CMV infection or
reactivation of latent virus.15 However, not all organs
from CMV(+) donors contain latent virus, and some studies report that
only 60% of CMV(+) donors transmit CMV even to seronegative recipients
at risk for primary infection.27 The substantially better
outcome for men compared with women receiving CMV(+) donor hearts
raises the question as to the source of observed sex differences.
CMV and Coronary Artery Disease
Much speculation as to the
potential mechanism(s) for interaction
between CMV infection, rejection, and coronary artery disease after
cardiac transplantation exists, though no causal relation has been
proved. One possible direct mechanism related to accelerated
atherosclerosis is via the induction of immunoglobulin Fc receptors on
CMV-infected endothelial cells. These receptors may interact with
granulocytes and lead to endothelial cell
damage.29 30
Other indirect theories include viral attachments to the blood vessel
surface, which then serve as haptens and activate the immune system.
This leads to damage of the neighboring vascular structures and
modification of blood vessel wall cell surface antigen
markers.22 In addition, the first five peptides of the CMV
show sequence homology and immunologic cross-reactivity with the HLA-DR
ß chain.31 This may lead to increased rejection after
transplantation in those infected with CMV by the cross-reactivity of
virus-directed antibodies with graft HLA
molecules.22 31
It is possible that the same mechanisms promoting accelerated
atherosclerosis and rejection after primary or reactivated CMV may also
contribute to graft loss and death after cardiac transplantation. In a
recent study of 8331 patients, Opelz and Wujciak32 found
that graft survival after cardiac transplantation is significantly
influenced by the extent of HLA compatibility. Since their group
studied a large population, a sizable number of patients, 128, received
grafts with no or only one HLA-A, -B or -DR mismatch. In our study we
did not find that HLA mismatching led to reduced survival after cardiac
transplantation; however, most of our patients, women and men, had
greater than three HLA mismatches.
OKT3 as a Risk Factor for Mortality
There are several reports
in the literature discussing outcome
after OKT3 induction on rejection,32 36 increased CMV
infection,33 34 35 37 and
survival33 38 in
patients undergoing cardiac or renal transplantation. OKT3 has not been
shown to significantly affect long-term rejection
rates.8 33 34 Three reports, however,
indicate that OKT3
is associated with increased CMV
infection33 35 37 38 in
small numbers of patients. The report by Johnson and
associates33 in cardiac transplantation indicates there is
no benefit on survival after OKT3 treatment. In this analysis, the
findings indicate OKT3 induction is associated with decreased survival
especially in recipients of organs from CMV(+) donors. There are
several confounding factors in our population that require
consideration in the analysis of the significance of OKT3 induction
as a risk factor for mortality after transplantation in women. First,
the population of women treated with OKT3 induction was small and
consisted of a sicker cohort (33% of women given OKT3 induction
therapy were UNOS status I versus 15% of women not given OKT3
induction therapy). Second, only beginning in 1989 was concurrent CMV
prophylaxis with immunoglobulin therapy used with OKT3 induction
therapy in female recipients of CMV(+) donors. Insufficient numbers
precluded analysis of OKT3 as a risk factor after CMV prophylaxis.
Given these limitations, it appears that OKT3 adversely affects
survival in women receiving a CMV(+) heart.
Possible Causes of Adverse Outcome in Women
Why do women fare
less well, reject more frequently, and die
sooner and in greater numbers than men of the same age? One possible
explanation could be related to differences in frequency of
autoimmune-mediated diseases in women.3 Women are at
higher risk for developing many autoimmune-mediated diseases including
systemic lupus erythematosus, idiopathic thrombocytopenic purpura,
primary biliary cirrhosis, rheumatoid arthritis, sarcoidosis, and
Graves' disease. Because sequence homology exists between CMV and
cardiac endothelium, an autoimmune mechanism whereby the body attacks
itself may lead to a more vigorous reaction in women than in men. Some
female patients carried the diagnosis of cardiomyopathy and may have
had an underlying autoimmune mechanism mediating involvement of the
native and transplanted heart. The issue of multiparity has also been
raised in an effort to account for poorer outcome after cardiac
transplantation via antibodies to foreign
material.6 12 13
However, in the present group of female patients, the majority were
multiparous, and no relation between number of pregnancies and outcome
was found.
Another factor considered was the influence of race on outcome. Of women in this study, 31% were nonwhite compared with 11% nonwhite men. A recent study of racial differences in cadaveric renal allograft survival revealed that decreased long-term survival of allografts is related not only to poor HLA matching but also to unfavorable socioeconomic factors and reduced compliance in black subjects.39 It is unlikely, however, that racial differences adequately address the disparate survival rates between women and men since, when stratified according to donor CMV status, survival of female recipients of CMV(-) donors, regardless of race, paralleled that of male recipients of CMV(-) donors.
Limitations
This retrospective study identifies the risk
factors for mortality
up to 36 months after transplantation. It does not determine if CMV
status is a marker or is causally related to outcome for rejection and
survival nor does it answer if CMV prophylaxis removes the apparent
increased risk for mortality in women treated with OKT3 induction. A
prospective study with a longer follow-up period might clarify the role
of CMV status in late mortality when accelerated coronary artery
disease occurs and may also reveal other factors responsible for poorer
outcome in women after transplantation. Although at this institution
during the study time period, CMV prophylaxis was not routinely
employed and OKT3 induction was selectively used in higher-risk
patients, conclusions regarding differences in outcome that are sex
dependent are valid.
Conclusions
CMV is a potential subfactor that identifies
increased risk for
mortality after cardiac transplantation. In summary: (1) Women are at
risk for reduced actuarial survival up to 3 years after cardiac
transplantation. (2) Univariate analysis shows that women are
selectively at risk for death when receiving hearts from CMV(+) donors
and after receiving OKT3 induction therapy. (3) Multivariate
analysis reveals that women are at even greater risk for death when
receiving hearts from CMV(+) donors in conjunction with OKT3 induction
therapy. (4) In the absence of OKT3 use, the greatest risk of death
occurs in CMV(-) women transplanted with CMV(+) donor hearts. (5)
When
survival curves of women and men were compared, factors that influenced
survival in women did not appear to be problematic in men.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received April 29, 1994; revision received August 5, 1999; accepted September 23, 1994.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. A. Fox and N. A. Nussmeier Does Gender Influence the Likelihood or Types of Complications Following Cardiac Surgery? Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2004; 8(4): 283 - 295. [Abstract] [PDF] |
||||
![]() |
S. Shumway Transplant and ventricular assist devices: Gender differences in application and implementation J. Thorac. Cardiovasc. Surg., May 1, 2004; 127(5): 1253 - 1255. [Full Text] [PDF] |
||||
![]() |
K. Lietz, R. John, A. Kocher, M. Schuster, D. M. Mancini, N. M. Edwards, and S. Itescu Increased Prevalence of Autoimmune Phenomena and Greater Risk for Alloreactivity in Female Heart Transplant Recipients Circulation, September 18, 2001; 104(90001): I-177 - 183. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Chemla, E. Aptecar, J.-L. Hebert, C. Coirault, D. Loisance, Y. Lecarpentier, and A. Nitenberg Short-term variability of pulse pressure and systolic and diastolic time in heart transplant recipients Am J Physiol Heart Circ Physiol, July 1, 2000; 279(1): H122 - H129. [Abstract] [Full Text] [PDF] |
||||
![]() |
More Women Than Men Die After Heart Transplant Journal Watch Cardiology, May 1, 1995; 1995(501): 17 - 17. [Full Text] |
||||
![]() |
MORTALITY WORSE FOR WOMEN AFTER CARDIAC TRANSPLANTATION Journal Watch (General), March 14, 1995; 1995(314): 2 - 2. [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |