- Email: [email protected]
Marital status improves survival after orthotopic heart transplantation
http://www.jhltonline.org
Marital status improves survival after orthotopic heart transplantation Vernissia Tam, BS,a George J. Arnaoutakis, MD,a Timothy J. George, MD,a Stuart D. Russell, MD,b Christian A. Merlo, MD, MPH,c,d John V. Conte, MD,a William A. Baumgartner, MD,a and Ashish S. Shah, MDa From the aDivision of Cardiac Surgery, the bDivision of Cardiology, the cDivision of Pulmonary and Critical Care Medicine, and the d Bloomberg School of Public Health, The Johns Hopkins Medical Institutions, Baltimore, Maryland
KEYWORDS: Heart transplantation; social support; marital status; non-adherence
BACKGROUND: Large national registries lack information on social support, which is increasingly recognized as an important factor associated with improved outcomes after solid-organ transplantation. We examined our institutional database to identify social factors associated with improved outcomes after orthotopic heart transplantation (OHT). METHODS: Outcomes of OHT patients from 1995 to 2010 at our institution were retrospectively reviewed. Clinical data and social information were extracted from medical records. Patients were stratified by marital status at time of OHT listing. The examined outcome was 5-year survival, excluding deaths within 60 days, modeled using the Kaplan-Meier method. A Cox multivariable hazard regression model was constructed to assess the effect on 5-year survival. RESULTS: Of 260 OHT recipients, 176 (68%) were men. Mean age was 49 ⫾ 12 years and mean body mass index was 26.8 ⫾ 5.0 kg/m2. At the time of OHT listing, 175 patients (68%) were married. Before OHT, 25% were supported with ventricular assist devices and 17% were in the intensive care unit. Conditional Kaplan-Meier analysis revealed improved 5-year survival for married patients (84%) compared with unmarried patients (69%). After risk-adjustment with Cox analysis, being married improved 5-year survival (hazard ratio [HR], 0.47; 95% confidence interval [CI], 0.23– 0.97; p ⫽ 0.042) and also improved 1-year survival (p ⫽ 0.02). Other social support variables (children, grandchildren, living arrangements, education, race, employment status) were not associated with mortality. CONCLUSIONS: Married patients have improved survival after OHT compared with unmarried patients. Being married confers a powerful 5-year survival advantage after OHT. This benefit appears mediated by an improvement in survival during the first post-transplant year. J Heart Lung Transplant 2011;30:1389 –94 © 2011 International Society for Heart and Lung Transplantation. All rights reserved.
Optimizing survival after orthotopic heart transplantation (OHT) necessitates a refined screening process for candidacy. Because caregivers are highly involved in nearly every aspect of post-operative care, social support is increasingly recognized as an important factor associated with improved outcomes after solid-organ transplantation.1– 4 In the past decade, research investigating the effect of social factors on recovery after OHT has been limited. Specifically, there is lack of data Reprint requests: Ashish S. Shah, MD, Assistant Professor of Surgery, Division of Cardiac Surgery, The Johns Hopkins Hospital, Blalock 618, 600 N Wolfe St, Baltimore, MD 21287. Telephone: 410-502-3900. Fax: 410-955-3809. E-mail address: [email protected]
describing long-term survival after OHT in patients with varying levels of social support. The effect of social factors on survival after OHT, such as marital status, living arrangements, and family composition, has not been thoroughly investigated. This is partly because large national registries lack social data; therefore, we examined our institutional database to test the hypothesis that strong social support is associated with improved outcomes after OHT.
Methods This retrospective review of OHT patients at the Johns Hopkins Hospital was approved by the Investigational Review Board.
1053-2498/$ -see front matter © 2011 International Society for Heart and Lung Transplantation. All rights reserved. doi:10.1016/j.healun.2011.07.020
1390
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011
Patient data Patients who underwent OHT between January 1995 and February 2010 were included. Relevant clinical information was extracted from TeleResults, a prospectively collected institutional database for transplant recipients, and the electronic medical record.
Clinical variables Clinical variables available in the database included body mass index (BMI), transplant diagnosis, time on the waiting list, hypertension, diabetes mellitus, serum creatinine, pulmonary hemodynamics, cardiac output, need for mechanical circulatory support (ventricular assist device, intra-aortic balloon pump, or extracorporeal membrane oxygenation), mechanical ventilation, cytomegalovirus (CMV) mismatch, human leukocyte antigen (HLA) mismatch, and allograft ischemic time.
Sociodemographic variables Demographic variables included age, sex, and race. Distance from transplant center was extracted from the database. History of mental health problems included depression, bipolar disorder, or anxiety disorder. Remote history of substance abuse (smoking, alcohol, cocaine, and/or intravenous drugs) was also recorded. This study defined social factors as those variables consistently present in the social history section of the social worker assessment of transplant candidacy. During the study period, all OHT recipients had undergone a standardized evaluation by members of the heart transplant social work team. These variables included marital status, family composition (spouse, children, and/or grandchildren), living situation (living alone, living with a spouse, and/or with children), and education level. Marital status was usually reported by the social worker as married, single, divorced, or widowed. There were 6 instances when the marital status as reported by the social worker did not fall into one of these discrete categories (i.e. living with a stable partner), and these patients were included in the subset of married patients.
Outcomes Survival status was available in the electronic database and crossreferenced using the Social Security Death Index. Examined outcomes were 5-year survival and time to first episode of drug-treated rejection. Rejection information was collected from biopsy notes from the electronic medical records and defined as any rejection episode requiring treatment.
power of the model. As the model was developed with case-wise deletion, covariates with greater than 15% missing data were not included. To explore an effect modification between various social factors and the outcomes, interaction terms between social factors were tested in the multivariable model. There were no significant tests of interaction, and therefore, no interaction term was left in the final multivariable model. Coefficients are presented with 95% confidence intervals (CI). The final Cox risk-adjusted multivariate model for 5-year survival included age at OHT, marital status at time of listing, number of children, diabetes, BMI, baseline creatinine, and CMV mismatch. The final Cox risk-adjusted multivariate model for freedom from rejection included age at OHT, marital status at time of listing, number of children, BMI, HLA mismatch, education, and travel time. Survival at 5 years was estimated using the Kaplan-Meier method, and the log-rank test compared survival stratified by marital status at the time of OHT listing (married vs non-married). Deaths within 60 days of transplant were likely due to transplant procedure or graft quality, and therefore, a conditional analysis excluding deaths within 60 days was performed.5 Values of p ⱕ 0.05 were considered significant for all tests. Analysis was performed using STATA 9.2 software (StataCorp, College Station, TX).
Results Baseline characteristics and sociodemographic variables Between January 1995 and February 2010, 260 patients (175 men, 68%), who were a mean age of 49 ⫾ 12 years, received OHT at our institution. Baseline and sociodemographic variables are summarized in Table 1. The racial distribution consisted of 195 (75%) white, 52 (20%) black, and 13 (5%) other. The most common indication for transplant was idiopathic cardiomyopathy (39%). Average baseline creatinine was 1.3 ⫾ 0.6 mg/ml. At the time of OHT listing, 175 patients (68%) were married, and 21 (8%) were living alone. Patients had an average of 1.9 ⫾ 1.6 children, 1.4 ⫾ 2.4 grandchildren, and 2.3 ⫾ 2.0 siblings. The average travel time to our institution was 1.6 ⫾ 1.2 hours.
Outcomes and survival Statistical analysis A Cox multivariable proportional hazards regression model was constructed to determine the relative hazard of individual variables on 5-year survival and freedom from the first episode of drugtreated rejection. Deaths within the first 60 days were likely due to transplant procedure or graft quality and were excluded from Cox regression analysis for both outcomes. We included variables associated with survival or rejection on an exploratory univariate analysis (p ⬍ 0.2), those with biologic plausibility, or those described in previously published studies in a forward and backward stepwise fashion into the model. The likelihood ratio test and Akaike’s information criterion were applied to a nested model approach to identify which covariates increase the explanatory
Mean follow-up time was 68 ⫾ 52 months. Twenty-eight patients (10%) died within 60 days and were excluded from Cox regression analyses. After adjustment with a Cox proportional hazards regression model, being married improved 5-year survival (hazard ratio [HR], 0.47; 95% CI, 0.23– 0.97; p ⫽ 0.042). Meanwhile, higher baseline creatinine was associated with worse 5-year survival (HR, 1.83; 95% CI, 1.19 –2.81, p ⫽ 0.006; Table 2). No significant associations were found for the variables of age at OHT, number of children, diabetes, BMI, or CMV mismatch. The remaining social variables were not associated with longterm mortality on univariate analysis or did not increase the
Tam et al.
Marital status improves survival after OHT
Table 1
Baseline Characteristics and Social Demographics
Variable Total patients, N Age, years Male sex Race White Black Other Body mass index, kg/m2 Diagnosis Idiopathic cardiomyopathy Ischemic Hypertrophic Othera Diabetes Baseline creatinine, mg/dl Pre-operative Inotropes Ventricular assist device Intensive care unit Time on waiting list, days Ischemic time, minutes Married Family composition Children, No. Grandchildren, No. Siblings, No. Living alone Travel time, hours Education ⬎ high school Employed History of Smoking Substance abuse Mental health disorder
Mean ⫾ SD (range) or No. (%) 260 49 ⫾ 12 (18–70) 176 (68) 195 (75) 52 (20) 13 (5) 26.8 ⫾ 5.0 (16.6–43.4) 101 (39) 83 (32) 47 (18) 29 (11) 68 (26) 1.3 ⫾ 0.6 (0.5–4.5) 60 (26) 62 (25) 40 (17) 331 ⫾ 353 (3–1,979) 183 ⫾ 59 (71–359) 175 (68) 1.9 ⫾ 1.6 1.4 ⫾ 2.4 2.3 ⫾ 2.0 21 (8) 1.6 ⫾ 1.2 125 (48) 85 (33)
(0–11) (0–10) (0–10) (0.5–5.5)
135 (52) 28 (11) 44 (17)
SD, Standard deviation. a Familial, post-partum, congenital.
explanatory power of the multivariable model and, therefore, were not included in the risk-adjusted model. Conditional Kaplan-Maier estimates (excluding deaths within 60 days) demonstrated a 5-year survival of 84% for married patients and 69% for unmarried patients (p ⬍ 0.01 by the log-rank test; Figure 1). Because the survival curves appear to diverge at 1 year, the log-rank test was performed at 1 year and demonstrated a statistically significant survival improvement for married patients (p ⫽ 0.01). We performed a sub-group analysis to examine Kaplan-Meier 5-year survival estimates among married patients only, stratified by sex (Figure 2). Although there was a trend in favor of improved survival among married men, this did not reach significance (p ⫽ 0.2). Median time to the first episode of drug-treated rejection was 4 months (interquartile range, 1–12 months). Cox proportional hazards regression analysis (excluding deaths within 60 days) revealed increasing age was protective against rejection (HR, 0.98; 95% CI, 0.96 – 0.99; p ⫽
1391 0.004). Meanwhile, greater travel time to the transplant center was associated with poorer freedom from rejection (HR, 1.16; 95% CI, 1.003–1.35, p ⫽ 0.046; Table 3). No significant associations were found for the variables of marital status, number of children, BMI, HLA mismatch, or education.
Discussion This study assessed the effect of social support on survival at 1 and 5 years after OHT. Owing to the lack of social support information in large national registries, we analyzed our institutional database of 260 patients who received OHT between 1995 and 2010. Patients who were married at the time of the pre-transplant evaluation had a 15% absolute improvement in 5-year survival after OHT. A significant survival advantage for married patients was evident at 1-year. Thus, these results suggest that the absolute improvement in 5-year survival for married patients is likely attributable to a survival benefit that is conferred during the first post-transplant year. Consistent with previous studies, we also found poor kidney function, as reflected by a high creatinine level, was associated with poor survival after OHT.6 Increased distance from the transplant center was associated with a shorter time to the first episode of drugtreated rejection.
Social support We speculate that spouses provide beneficial support to the overall health of OHT recipients, including the management of comorbidities. Our results are consistent with those of Dobbels et al,2 who studied pre-transplant psychosocial predictors of post-transplant medical compliance and found that lack of partnership was the only significant predictor of graft loss between 6 and 12 months after transplant. Our results are consistent with their findings: the survival advantage in our series reached significance at 1 year and persisted at 5 years. In addition, Chacko et al7 used selfreported psychometric tests to examine 94 OHT patients and found that the most reliable predictors of survival were self-assessed measures of coping and social support. A sub-group survival analysis among married patients only, stratified by sex, found no statistically significant difference, although a trend could be appreciated. A true difference might have been detected in a larger sample of patients. This may represent pure biologic differences but also may reflect a differential advantage of marital support based on the sex of the recipient. Many studies have documented the psychiatric distress that accompanies organ replacement. Caregivers provide assistance with transportation and medical compliance, economic benefits, and emotional support.8,9 The absence of social support has been considered an absolute and relative contraindication to transplant by 38% and 81% of experts in the OHT field, respectively.10 Marriage may be a surrogate for forms of support that are difficult to quantify but extend
1392 Table 2
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011 Cox Multivariable Hazard Regression Model—5-Year Survival Univariate analysis
Multivariate analysis
Variable
HR (95% CI)
p-value
HR (95% CI
p-value
Age Married Children Diabetes BMI Creatinine CMV mismatch
0.98 0.54 0.89 1.4 1.03 1.4 1.7
0.098 0.039 0.26 0.25 0.39 0.12 0.11
0.98 0.47 0.98 1.1 1.03 1.8 1.5
0.28 0.042a 0.87 0.80 0.29 0.006a 0.25
(0.96–1.003) (0.30–0.97) (0.73–1.1) (0.77–2.7) (0.97–1.09) (0.91–2.3) (0.88–3.2)
(0.96–1.01) (0.23–0.97) (0.76–1.3) (0.49–2.5) (0.97–1.1) (1.2–2.8) (0.75–2.9)
BMI, body mass index; CI, confidence interval; CMV, cytomegalovirus; HR, hazard ratio. a p-value ⬍ 0.05 significance level.
beyond the physical responsibilities of a caregiver. As a result, married OHT recipients self-report improved selfesteem, satisfaction with life, and well-being.11 Given the documented psychiatric distress that accompanies organ replacement and the demand for caregiver support throughout the OHT process, married recipients matched our expectations for achieving greater survival at 1 and 5 years.1,12 Shapiro et al4 documented the level of pre-transplant social support, as assessed by a psychiatrist, in a prospective study of 125 OHT recipients. Contrary to our results, no association was found between social support and survival.4 However, the mean follow-up time was 13.8 ⫾ 9.9 months, and that study may have been insufficiently powered to detect survival differences, given the 1-year survival rate of 88% for OHT recipients in the United States.13 We observed a survival improvement at 1 year, which may reflect the larger sample size.
Non-adherence Non-adherence after OHT is associated with graft loss, rejection, and death.3 Although one might expect transplant recipients to be a highly motivated group, medical nonadherence with immunosuppressive drugs for OHT recipients may be as high as 25%.1 Dew et al14 proposed that the strongest predictors of non-adherence after OHT were psychosocial risk factors, including the quality of family rela-
Figure 1 Kaplan-Meier estimates of 5-year survival stratified by marital status (p-value determined by Cox-Mantel log-rank test).
tionships, whereas non-compliant kidney recipients were more likely to be single.14 Unfortunately, our database lacks surrogate measures for compliance; however, we surmise that being married may confer a survival benefit through increased compliance with management of all comorbidities. Given this assumption, we would have expected marital status to also confer a protective effect against drugtreated rejection. Although we did observe a trend in favor of marital status protecting against rejection on univariate analysis, this did not bear out in risk-adjusted multivariable analysis. Our sample may lack the power to detect a true difference, and further study is warranted. In addition, our survival analysis included all-cause mortality, and graft-related causes of death in our institutional experience account for 40% of deaths (data not shown). Being married likely confers a beneficial effect in ways that are difficult to measure and not simply by improved compliance with anti-rejection medications. Our study found that increasing age was associated with a greater freedom from rejection. This is consistent with past studies that found an increased incidence of non-adherence in renal, heart, and liver transplant patients who were aged younger than 20 years.3,15 Our findings may be explained by greater responsibility upheld by older OHT recipients or waning immunity with advancing age.16,17 Every hour increase in travel time to our transplant center was associated with a 16% decrease in freedom from rejection. We propose that increased distance from the
Figure 2 Kaplan-Meier estimates of 5-year survival stratified by sex among married patients (p-value determined by Cox-Mantel log-rank test).
Tam et al. Table 3
Marital status improves survival after OHT
1393
Cox Multivariable Hazard Regression Model—Freedom From Rejection Univariate analysis
Multivariate analysis
Variable
HR (95% CI)
p-value
HR (95% CI)
p-value
Age Married Children BMI HLA mismatch Education ⬎ HS Travel time
0.98 0.77 0.69 1.02 1.15 0.84 1.08
0.001 0.18 0.076 0.17 0.58 0.36 0.27
0.98 0.95 1.07 1.02 0.95 0.82 1.2
0.004a 0.82 0.81 0.25 0.84 0.29 0.046a
(0.96–0.99) (0.53–1.12) (0.46–1.04) (0.99–1.06) (0.68–1.96) (0.60–1.20) (0.94–1.3)
(0.96–0.99) (0.62–1.5) (0.61–1.9) (0.99–1.06) (0.55–1.6) (0.57–1.2) (1.002–1.4)
BMI, body mass index; CI, confidence interval; HLA, human leukocyte antigen; HR, hazard ratio; HS, high school. a p-value ⬍ 0.05 significance level.
transplant center may be associated with medical and appointment non-adherence, decreased accessibility to specialized health care, and/or decreased likelihood to refill prescriptions. In the De Geest et al18 study of 101 OHT recipients, late acute rejection episodes occurred more often in the group of appointment non-compliers, who were also younger, less likely to live with a stable partner, were more depressed, and had more drug holidays.18 In contrast with our findings, DeGeest et al did not show that distance from transplant center was linked with rejection, and additional studies are needed to clarify this issue. Many background health and demographic variables, including cardiac diagnosis and race, showed no significant influence on post-OHT survival or rejection, despite past findings.13 Previous studies have demonstrated that recipient age, race, pre-operative alcohol abuse, smoking, personality disorders, increased BMI, and higher education were associated with poor compliance after OHT.2– 4,19,20 However, these variables were not associated with death or rejection in our study. It is likely that with larger sample size we would have had increased power to detect more subtle differences and would have been able to construct our regression analysis more consistent with International Society for Heart and Lung Transplantation (ISHLT) Registry. It is important to note that our regression analysis excluded many of the clinical variables that contribute to early deaths. Our study has a few limitations: Our analysis lacks certain socioeconomic variables, such as actual income and insurance type, that may be confounding factors in accessing medical care. Marital status is not a perfect surrogate for social support, which is only one potential mechanism that may explain differences in survival outcome. However, we attempted to control for all clinical and social variables available through our database and the electronic medical record. All social variables were collected as recorded at the time of the pre-OHT evaluation. Therefore, our analysis did not account for changes in variables that might have occurred after the evaluation or even after the transplantation. Our study cannot address the effect of changing social support dynamics post-OHT compared with pre-OHT status.
Owing to the retrospective method of our study using electronic medical records, we were unable to assess the quality of the relationship between the recipient and the caregiver. Specifically, we were unable to address the quality of the marital relationship that has been previously described using various tools such as the Dyadic Adjustment Scale.21 Although we assume that legal marriage implies strong social support, patients have reported that potential sources of support were lost in the face of OHT.22 Alternatively, the lack of a legal spouse does not preclude strong social support from other family members and friends.12 The relatively small sample size may have limited our ability to detect differences in sub-group analysis, as well as when testing interaction terms. Although married men did not demonstrate a statistically significant survival advantage, the power to detect significant differences was lower in this sub-group analysis. Despite no statistically significant difference, inspection of the Kaplan-Meier curves does reveal a prominent absolute difference. When examining the cohort collectively with men and women combined, the sample size had sufficient power to detect a difference. If national registries such as the United Network for Organ Sharing request additional social support information, larger-scale studies can be conducted to further address these issues. In conclusion, this study demonstrated that being married confers a 1-year survival advantage after OHT that persists at 5 years. These results identify high-risk patients who warrant additional attention by transplant teams to reduce mortality, morbidity, and non-adherence. Additional study is warranted to further explain these results.
Disclosure statement Presented at the Concurrent Session of The International Society for Heart and Lung Transplantation’s Thirty-first Annual Meeting and Scientific Sessions, San Diego, California, Saturday, April 14, 2011. The authors thank Jennifer Neeley for her assistance in managing the heart transplant database at the Johns Hopkins Hospital. This research was supported in part by the National Institutes of Health Grant 1T32CA126607⫺01A2 to Dr Arnaoutakis.
1394
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011
Dr Arnaoutakis is the Irene Piccinini Investigator in Cardiac Surgery, and Dr George is the Hugh R. Sharp Cardiac Surgery Research Fellow. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
References 1. Dew MA, Roth LH, Thompson ME, Kormos RL, Griffith BP. Medical compliance and its predictors in the first year after heart transplantation. J Heart Lung Transplant 1996;15:631-45. 2. Dobbels F, Vanhaecke J, Dupont L, et al. Pretransplant predictors of posttransplant adherence and clinical outcome: an evidence base for pretransplant psychosocial screening. Transplantation 2009;87:1497-504. 3. Schweizer RT, Rovelli M, Palmeri D, Vossler E, Hull D, Bartus S. Noncompliance in organ transplant recipients. Transplantation 1990; 49:374-7. 4. Shapiro PA, Williams DL, Foray AT, Gelman IS, Wukich N, Sciacca R. Psychosocial evaluation and prediction of compliance problems and morbidity after heart transplantation. Transplantation 1995;60:1462-6. 5. Lindenfeld J, Miller GG, Shakar SF, et al. Drug therapy in the heart transplant recipient: part II: immunosuppressive drugs. Circulation 2004;110:3858-65. 6. Boyle JM, Moualla S, Arrigain S, et al. Risks and outcomes of acute kidney injury requiring dialysis after cardiac transplantation. Am J Kidney Dis 2006;48:787-96. 7. Chacko RC, Harper RG, Gotto J, Young J. Psychiatric interview and psychometric predictors of cardiac transplant survival. Am J Psychiatry 1996;153:1607-12. 8. Muirhead J, Meyerowitz BE, Leedham B, Eastburn TE, Merrill WH, Frist WH. Quality of life and coping in patients awaiting heart transplantation. J Heart Lung Transplant 1992;11:265-71; discussion: 271–2. 9. Singh TP, Gauvreau K, Bastardi HJ, Blume ED, Mayer JE. Socioeconomic position and graft failure in pediatric heart transplant recipients. Circ Heart Fail 2009;2:160-5.
10. Sirri L, Magelli C, Grandi S. Predictors of perceived social support in long-term survivors of cardiac transplant: the role of psychological distress, quality of life, demographic characteristics and clinical course. Psychol Health 2011;26:77-94. 11. Bohachick P, Taylor MV, Sereika S, Reeder S, Anton BB. Social support, personal control, and psychosocial recovery following heart transplantation. Clin Nurs Res 2002;11:34-51. 12. Bunzel B, Wollenek G. Heart transplantation: are there psychosocial predictors for clinical success of surgery? Thorac Cardiovasc Surg 1994;42:103-7. 13. Allen JG, Weiss ES, Arnaoutakis GJ, et al. The impact of race on survival after heart transplantation: an analysis of more than 20,000 patients. Ann Thorac Surg 2010;89:1956-63; discussion: 1963– 4. 14. Matas M, Staley D, Griffin W. A profile of the noncompliant patient: a thirty-month review of outpatient psychiatry referrals. Gen Hosp Psychiatry 1992;14:124-30. 15. Beck DE, Fennell RS, Yost RL, Robinson JD, Geary D, Richards GA. Evaluation of an educational program on compliance with medication regimens in pediatric patients with renal transplants. J Pediatr 1980; 96:1094-7. 16. Flurkey K, Stadecker M, Miller RA. Memory T lymphocyte hyporesponsiveness to non-cognate stimuli: a key factor in age-related immunodeficiency. Eur J Immunol 1992;22:931-5. 17. Pawelec G, Barnett Y, Forsey R, et al. T cells and aging, January 2002 update. Front Biosci 2002;7:d1056-183. 18. De Geest S, Dobbels F, Martin S, Willems K, Vanhaecke J. Clinical risk associated with appointment noncompliance in heart transplant recipients. Prog Transplant 2000;10:162-8. 19. Grady KL, Costanzo MR, Fisher S, Koch D. Preoperative obesity is associated with decreased survival after heart transplantation. J Heart Lung Transplant 1996;15:863-71. 20. Nagele H, Kalmar P, Rodiger W, Stubbe HM. Smoking after heart transplantation: an underestimated hazard? Eur J Cardiothorac Surg 1997;12:70-4. 21. Lawrence E, Barry RA, Brock RL, et al. The Relationship Quality Interview: evidence of reliability, convergent and divergent validity, and incremental utility. Psychol Assess 2011;23:44-63. 22. Kaba E, Thompson DR, Burnard P. Coping after heart transplantation: a descriptive study of heart transplant recipients’ methods of coping. J Adv Nurs 2000;32:930-6.
Marital status improves survival after orthotopic heart transplantation Vernissia Tam, BS,a George J. Arnaoutakis, MD,a Timothy J. George, MD,a Stuart D. Russell, MD,b Christian A. Merlo, MD, MPH,c,d John V. Conte, MD,a William A. Baumgartner, MD,a and Ashish S. Shah, MDa From the aDivision of Cardiac Surgery, the bDivision of Cardiology, the cDivision of Pulmonary and Critical Care Medicine, and the d Bloomberg School of Public Health, The Johns Hopkins Medical Institutions, Baltimore, Maryland
KEYWORDS: Heart transplantation; social support; marital status; non-adherence
BACKGROUND: Large national registries lack information on social support, which is increasingly recognized as an important factor associated with improved outcomes after solid-organ transplantation. We examined our institutional database to identify social factors associated with improved outcomes after orthotopic heart transplantation (OHT). METHODS: Outcomes of OHT patients from 1995 to 2010 at our institution were retrospectively reviewed. Clinical data and social information were extracted from medical records. Patients were stratified by marital status at time of OHT listing. The examined outcome was 5-year survival, excluding deaths within 60 days, modeled using the Kaplan-Meier method. A Cox multivariable hazard regression model was constructed to assess the effect on 5-year survival. RESULTS: Of 260 OHT recipients, 176 (68%) were men. Mean age was 49 ⫾ 12 years and mean body mass index was 26.8 ⫾ 5.0 kg/m2. At the time of OHT listing, 175 patients (68%) were married. Before OHT, 25% were supported with ventricular assist devices and 17% were in the intensive care unit. Conditional Kaplan-Meier analysis revealed improved 5-year survival for married patients (84%) compared with unmarried patients (69%). After risk-adjustment with Cox analysis, being married improved 5-year survival (hazard ratio [HR], 0.47; 95% confidence interval [CI], 0.23– 0.97; p ⫽ 0.042) and also improved 1-year survival (p ⫽ 0.02). Other social support variables (children, grandchildren, living arrangements, education, race, employment status) were not associated with mortality. CONCLUSIONS: Married patients have improved survival after OHT compared with unmarried patients. Being married confers a powerful 5-year survival advantage after OHT. This benefit appears mediated by an improvement in survival during the first post-transplant year. J Heart Lung Transplant 2011;30:1389 –94 © 2011 International Society for Heart and Lung Transplantation. All rights reserved.
Optimizing survival after orthotopic heart transplantation (OHT) necessitates a refined screening process for candidacy. Because caregivers are highly involved in nearly every aspect of post-operative care, social support is increasingly recognized as an important factor associated with improved outcomes after solid-organ transplantation.1– 4 In the past decade, research investigating the effect of social factors on recovery after OHT has been limited. Specifically, there is lack of data Reprint requests: Ashish S. Shah, MD, Assistant Professor of Surgery, Division of Cardiac Surgery, The Johns Hopkins Hospital, Blalock 618, 600 N Wolfe St, Baltimore, MD 21287. Telephone: 410-502-3900. Fax: 410-955-3809. E-mail address: [email protected]
describing long-term survival after OHT in patients with varying levels of social support. The effect of social factors on survival after OHT, such as marital status, living arrangements, and family composition, has not been thoroughly investigated. This is partly because large national registries lack social data; therefore, we examined our institutional database to test the hypothesis that strong social support is associated with improved outcomes after OHT.
Methods This retrospective review of OHT patients at the Johns Hopkins Hospital was approved by the Investigational Review Board.
1053-2498/$ -see front matter © 2011 International Society for Heart and Lung Transplantation. All rights reserved. doi:10.1016/j.healun.2011.07.020
1390
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011
Patient data Patients who underwent OHT between January 1995 and February 2010 were included. Relevant clinical information was extracted from TeleResults, a prospectively collected institutional database for transplant recipients, and the electronic medical record.
Clinical variables Clinical variables available in the database included body mass index (BMI), transplant diagnosis, time on the waiting list, hypertension, diabetes mellitus, serum creatinine, pulmonary hemodynamics, cardiac output, need for mechanical circulatory support (ventricular assist device, intra-aortic balloon pump, or extracorporeal membrane oxygenation), mechanical ventilation, cytomegalovirus (CMV) mismatch, human leukocyte antigen (HLA) mismatch, and allograft ischemic time.
Sociodemographic variables Demographic variables included age, sex, and race. Distance from transplant center was extracted from the database. History of mental health problems included depression, bipolar disorder, or anxiety disorder. Remote history of substance abuse (smoking, alcohol, cocaine, and/or intravenous drugs) was also recorded. This study defined social factors as those variables consistently present in the social history section of the social worker assessment of transplant candidacy. During the study period, all OHT recipients had undergone a standardized evaluation by members of the heart transplant social work team. These variables included marital status, family composition (spouse, children, and/or grandchildren), living situation (living alone, living with a spouse, and/or with children), and education level. Marital status was usually reported by the social worker as married, single, divorced, or widowed. There were 6 instances when the marital status as reported by the social worker did not fall into one of these discrete categories (i.e. living with a stable partner), and these patients were included in the subset of married patients.
Outcomes Survival status was available in the electronic database and crossreferenced using the Social Security Death Index. Examined outcomes were 5-year survival and time to first episode of drug-treated rejection. Rejection information was collected from biopsy notes from the electronic medical records and defined as any rejection episode requiring treatment.
power of the model. As the model was developed with case-wise deletion, covariates with greater than 15% missing data were not included. To explore an effect modification between various social factors and the outcomes, interaction terms between social factors were tested in the multivariable model. There were no significant tests of interaction, and therefore, no interaction term was left in the final multivariable model. Coefficients are presented with 95% confidence intervals (CI). The final Cox risk-adjusted multivariate model for 5-year survival included age at OHT, marital status at time of listing, number of children, diabetes, BMI, baseline creatinine, and CMV mismatch. The final Cox risk-adjusted multivariate model for freedom from rejection included age at OHT, marital status at time of listing, number of children, BMI, HLA mismatch, education, and travel time. Survival at 5 years was estimated using the Kaplan-Meier method, and the log-rank test compared survival stratified by marital status at the time of OHT listing (married vs non-married). Deaths within 60 days of transplant were likely due to transplant procedure or graft quality, and therefore, a conditional analysis excluding deaths within 60 days was performed.5 Values of p ⱕ 0.05 were considered significant for all tests. Analysis was performed using STATA 9.2 software (StataCorp, College Station, TX).
Results Baseline characteristics and sociodemographic variables Between January 1995 and February 2010, 260 patients (175 men, 68%), who were a mean age of 49 ⫾ 12 years, received OHT at our institution. Baseline and sociodemographic variables are summarized in Table 1. The racial distribution consisted of 195 (75%) white, 52 (20%) black, and 13 (5%) other. The most common indication for transplant was idiopathic cardiomyopathy (39%). Average baseline creatinine was 1.3 ⫾ 0.6 mg/ml. At the time of OHT listing, 175 patients (68%) were married, and 21 (8%) were living alone. Patients had an average of 1.9 ⫾ 1.6 children, 1.4 ⫾ 2.4 grandchildren, and 2.3 ⫾ 2.0 siblings. The average travel time to our institution was 1.6 ⫾ 1.2 hours.
Outcomes and survival Statistical analysis A Cox multivariable proportional hazards regression model was constructed to determine the relative hazard of individual variables on 5-year survival and freedom from the first episode of drugtreated rejection. Deaths within the first 60 days were likely due to transplant procedure or graft quality and were excluded from Cox regression analysis for both outcomes. We included variables associated with survival or rejection on an exploratory univariate analysis (p ⬍ 0.2), those with biologic plausibility, or those described in previously published studies in a forward and backward stepwise fashion into the model. The likelihood ratio test and Akaike’s information criterion were applied to a nested model approach to identify which covariates increase the explanatory
Mean follow-up time was 68 ⫾ 52 months. Twenty-eight patients (10%) died within 60 days and were excluded from Cox regression analyses. After adjustment with a Cox proportional hazards regression model, being married improved 5-year survival (hazard ratio [HR], 0.47; 95% CI, 0.23– 0.97; p ⫽ 0.042). Meanwhile, higher baseline creatinine was associated with worse 5-year survival (HR, 1.83; 95% CI, 1.19 –2.81, p ⫽ 0.006; Table 2). No significant associations were found for the variables of age at OHT, number of children, diabetes, BMI, or CMV mismatch. The remaining social variables were not associated with longterm mortality on univariate analysis or did not increase the
Tam et al.
Marital status improves survival after OHT
Table 1
Baseline Characteristics and Social Demographics
Variable Total patients, N Age, years Male sex Race White Black Other Body mass index, kg/m2 Diagnosis Idiopathic cardiomyopathy Ischemic Hypertrophic Othera Diabetes Baseline creatinine, mg/dl Pre-operative Inotropes Ventricular assist device Intensive care unit Time on waiting list, days Ischemic time, minutes Married Family composition Children, No. Grandchildren, No. Siblings, No. Living alone Travel time, hours Education ⬎ high school Employed History of Smoking Substance abuse Mental health disorder
Mean ⫾ SD (range) or No. (%) 260 49 ⫾ 12 (18–70) 176 (68) 195 (75) 52 (20) 13 (5) 26.8 ⫾ 5.0 (16.6–43.4) 101 (39) 83 (32) 47 (18) 29 (11) 68 (26) 1.3 ⫾ 0.6 (0.5–4.5) 60 (26) 62 (25) 40 (17) 331 ⫾ 353 (3–1,979) 183 ⫾ 59 (71–359) 175 (68) 1.9 ⫾ 1.6 1.4 ⫾ 2.4 2.3 ⫾ 2.0 21 (8) 1.6 ⫾ 1.2 125 (48) 85 (33)
(0–11) (0–10) (0–10) (0.5–5.5)
135 (52) 28 (11) 44 (17)
SD, Standard deviation. a Familial, post-partum, congenital.
explanatory power of the multivariable model and, therefore, were not included in the risk-adjusted model. Conditional Kaplan-Maier estimates (excluding deaths within 60 days) demonstrated a 5-year survival of 84% for married patients and 69% for unmarried patients (p ⬍ 0.01 by the log-rank test; Figure 1). Because the survival curves appear to diverge at 1 year, the log-rank test was performed at 1 year and demonstrated a statistically significant survival improvement for married patients (p ⫽ 0.01). We performed a sub-group analysis to examine Kaplan-Meier 5-year survival estimates among married patients only, stratified by sex (Figure 2). Although there was a trend in favor of improved survival among married men, this did not reach significance (p ⫽ 0.2). Median time to the first episode of drug-treated rejection was 4 months (interquartile range, 1–12 months). Cox proportional hazards regression analysis (excluding deaths within 60 days) revealed increasing age was protective against rejection (HR, 0.98; 95% CI, 0.96 – 0.99; p ⫽
1391 0.004). Meanwhile, greater travel time to the transplant center was associated with poorer freedom from rejection (HR, 1.16; 95% CI, 1.003–1.35, p ⫽ 0.046; Table 3). No significant associations were found for the variables of marital status, number of children, BMI, HLA mismatch, or education.
Discussion This study assessed the effect of social support on survival at 1 and 5 years after OHT. Owing to the lack of social support information in large national registries, we analyzed our institutional database of 260 patients who received OHT between 1995 and 2010. Patients who were married at the time of the pre-transplant evaluation had a 15% absolute improvement in 5-year survival after OHT. A significant survival advantage for married patients was evident at 1-year. Thus, these results suggest that the absolute improvement in 5-year survival for married patients is likely attributable to a survival benefit that is conferred during the first post-transplant year. Consistent with previous studies, we also found poor kidney function, as reflected by a high creatinine level, was associated with poor survival after OHT.6 Increased distance from the transplant center was associated with a shorter time to the first episode of drugtreated rejection.
Social support We speculate that spouses provide beneficial support to the overall health of OHT recipients, including the management of comorbidities. Our results are consistent with those of Dobbels et al,2 who studied pre-transplant psychosocial predictors of post-transplant medical compliance and found that lack of partnership was the only significant predictor of graft loss between 6 and 12 months after transplant. Our results are consistent with their findings: the survival advantage in our series reached significance at 1 year and persisted at 5 years. In addition, Chacko et al7 used selfreported psychometric tests to examine 94 OHT patients and found that the most reliable predictors of survival were self-assessed measures of coping and social support. A sub-group survival analysis among married patients only, stratified by sex, found no statistically significant difference, although a trend could be appreciated. A true difference might have been detected in a larger sample of patients. This may represent pure biologic differences but also may reflect a differential advantage of marital support based on the sex of the recipient. Many studies have documented the psychiatric distress that accompanies organ replacement. Caregivers provide assistance with transportation and medical compliance, economic benefits, and emotional support.8,9 The absence of social support has been considered an absolute and relative contraindication to transplant by 38% and 81% of experts in the OHT field, respectively.10 Marriage may be a surrogate for forms of support that are difficult to quantify but extend
1392 Table 2
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011 Cox Multivariable Hazard Regression Model—5-Year Survival Univariate analysis
Multivariate analysis
Variable
HR (95% CI)
p-value
HR (95% CI
p-value
Age Married Children Diabetes BMI Creatinine CMV mismatch
0.98 0.54 0.89 1.4 1.03 1.4 1.7
0.098 0.039 0.26 0.25 0.39 0.12 0.11
0.98 0.47 0.98 1.1 1.03 1.8 1.5
0.28 0.042a 0.87 0.80 0.29 0.006a 0.25
(0.96–1.003) (0.30–0.97) (0.73–1.1) (0.77–2.7) (0.97–1.09) (0.91–2.3) (0.88–3.2)
(0.96–1.01) (0.23–0.97) (0.76–1.3) (0.49–2.5) (0.97–1.1) (1.2–2.8) (0.75–2.9)
BMI, body mass index; CI, confidence interval; CMV, cytomegalovirus; HR, hazard ratio. a p-value ⬍ 0.05 significance level.
beyond the physical responsibilities of a caregiver. As a result, married OHT recipients self-report improved selfesteem, satisfaction with life, and well-being.11 Given the documented psychiatric distress that accompanies organ replacement and the demand for caregiver support throughout the OHT process, married recipients matched our expectations for achieving greater survival at 1 and 5 years.1,12 Shapiro et al4 documented the level of pre-transplant social support, as assessed by a psychiatrist, in a prospective study of 125 OHT recipients. Contrary to our results, no association was found between social support and survival.4 However, the mean follow-up time was 13.8 ⫾ 9.9 months, and that study may have been insufficiently powered to detect survival differences, given the 1-year survival rate of 88% for OHT recipients in the United States.13 We observed a survival improvement at 1 year, which may reflect the larger sample size.
Non-adherence Non-adherence after OHT is associated with graft loss, rejection, and death.3 Although one might expect transplant recipients to be a highly motivated group, medical nonadherence with immunosuppressive drugs for OHT recipients may be as high as 25%.1 Dew et al14 proposed that the strongest predictors of non-adherence after OHT were psychosocial risk factors, including the quality of family rela-
Figure 1 Kaplan-Meier estimates of 5-year survival stratified by marital status (p-value determined by Cox-Mantel log-rank test).
tionships, whereas non-compliant kidney recipients were more likely to be single.14 Unfortunately, our database lacks surrogate measures for compliance; however, we surmise that being married may confer a survival benefit through increased compliance with management of all comorbidities. Given this assumption, we would have expected marital status to also confer a protective effect against drugtreated rejection. Although we did observe a trend in favor of marital status protecting against rejection on univariate analysis, this did not bear out in risk-adjusted multivariable analysis. Our sample may lack the power to detect a true difference, and further study is warranted. In addition, our survival analysis included all-cause mortality, and graft-related causes of death in our institutional experience account for 40% of deaths (data not shown). Being married likely confers a beneficial effect in ways that are difficult to measure and not simply by improved compliance with anti-rejection medications. Our study found that increasing age was associated with a greater freedom from rejection. This is consistent with past studies that found an increased incidence of non-adherence in renal, heart, and liver transplant patients who were aged younger than 20 years.3,15 Our findings may be explained by greater responsibility upheld by older OHT recipients or waning immunity with advancing age.16,17 Every hour increase in travel time to our transplant center was associated with a 16% decrease in freedom from rejection. We propose that increased distance from the
Figure 2 Kaplan-Meier estimates of 5-year survival stratified by sex among married patients (p-value determined by Cox-Mantel log-rank test).
Tam et al. Table 3
Marital status improves survival after OHT
1393
Cox Multivariable Hazard Regression Model—Freedom From Rejection Univariate analysis
Multivariate analysis
Variable
HR (95% CI)
p-value
HR (95% CI)
p-value
Age Married Children BMI HLA mismatch Education ⬎ HS Travel time
0.98 0.77 0.69 1.02 1.15 0.84 1.08
0.001 0.18 0.076 0.17 0.58 0.36 0.27
0.98 0.95 1.07 1.02 0.95 0.82 1.2
0.004a 0.82 0.81 0.25 0.84 0.29 0.046a
(0.96–0.99) (0.53–1.12) (0.46–1.04) (0.99–1.06) (0.68–1.96) (0.60–1.20) (0.94–1.3)
(0.96–0.99) (0.62–1.5) (0.61–1.9) (0.99–1.06) (0.55–1.6) (0.57–1.2) (1.002–1.4)
BMI, body mass index; CI, confidence interval; HLA, human leukocyte antigen; HR, hazard ratio; HS, high school. a p-value ⬍ 0.05 significance level.
transplant center may be associated with medical and appointment non-adherence, decreased accessibility to specialized health care, and/or decreased likelihood to refill prescriptions. In the De Geest et al18 study of 101 OHT recipients, late acute rejection episodes occurred more often in the group of appointment non-compliers, who were also younger, less likely to live with a stable partner, were more depressed, and had more drug holidays.18 In contrast with our findings, DeGeest et al did not show that distance from transplant center was linked with rejection, and additional studies are needed to clarify this issue. Many background health and demographic variables, including cardiac diagnosis and race, showed no significant influence on post-OHT survival or rejection, despite past findings.13 Previous studies have demonstrated that recipient age, race, pre-operative alcohol abuse, smoking, personality disorders, increased BMI, and higher education were associated with poor compliance after OHT.2– 4,19,20 However, these variables were not associated with death or rejection in our study. It is likely that with larger sample size we would have had increased power to detect more subtle differences and would have been able to construct our regression analysis more consistent with International Society for Heart and Lung Transplantation (ISHLT) Registry. It is important to note that our regression analysis excluded many of the clinical variables that contribute to early deaths. Our study has a few limitations: Our analysis lacks certain socioeconomic variables, such as actual income and insurance type, that may be confounding factors in accessing medical care. Marital status is not a perfect surrogate for social support, which is only one potential mechanism that may explain differences in survival outcome. However, we attempted to control for all clinical and social variables available through our database and the electronic medical record. All social variables were collected as recorded at the time of the pre-OHT evaluation. Therefore, our analysis did not account for changes in variables that might have occurred after the evaluation or even after the transplantation. Our study cannot address the effect of changing social support dynamics post-OHT compared with pre-OHT status.
Owing to the retrospective method of our study using electronic medical records, we were unable to assess the quality of the relationship between the recipient and the caregiver. Specifically, we were unable to address the quality of the marital relationship that has been previously described using various tools such as the Dyadic Adjustment Scale.21 Although we assume that legal marriage implies strong social support, patients have reported that potential sources of support were lost in the face of OHT.22 Alternatively, the lack of a legal spouse does not preclude strong social support from other family members and friends.12 The relatively small sample size may have limited our ability to detect differences in sub-group analysis, as well as when testing interaction terms. Although married men did not demonstrate a statistically significant survival advantage, the power to detect significant differences was lower in this sub-group analysis. Despite no statistically significant difference, inspection of the Kaplan-Meier curves does reveal a prominent absolute difference. When examining the cohort collectively with men and women combined, the sample size had sufficient power to detect a difference. If national registries such as the United Network for Organ Sharing request additional social support information, larger-scale studies can be conducted to further address these issues. In conclusion, this study demonstrated that being married confers a 1-year survival advantage after OHT that persists at 5 years. These results identify high-risk patients who warrant additional attention by transplant teams to reduce mortality, morbidity, and non-adherence. Additional study is warranted to further explain these results.
Disclosure statement Presented at the Concurrent Session of The International Society for Heart and Lung Transplantation’s Thirty-first Annual Meeting and Scientific Sessions, San Diego, California, Saturday, April 14, 2011. The authors thank Jennifer Neeley for her assistance in managing the heart transplant database at the Johns Hopkins Hospital. This research was supported in part by the National Institutes of Health Grant 1T32CA126607⫺01A2 to Dr Arnaoutakis.
1394
The Journal of Heart and Lung Transplantation, Vol 30, No 12, December 2011
Dr Arnaoutakis is the Irene Piccinini Investigator in Cardiac Surgery, and Dr George is the Hugh R. Sharp Cardiac Surgery Research Fellow. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
References 1. Dew MA, Roth LH, Thompson ME, Kormos RL, Griffith BP. Medical compliance and its predictors in the first year after heart transplantation. J Heart Lung Transplant 1996;15:631-45. 2. Dobbels F, Vanhaecke J, Dupont L, et al. Pretransplant predictors of posttransplant adherence and clinical outcome: an evidence base for pretransplant psychosocial screening. Transplantation 2009;87:1497-504. 3. Schweizer RT, Rovelli M, Palmeri D, Vossler E, Hull D, Bartus S. Noncompliance in organ transplant recipients. Transplantation 1990; 49:374-7. 4. Shapiro PA, Williams DL, Foray AT, Gelman IS, Wukich N, Sciacca R. Psychosocial evaluation and prediction of compliance problems and morbidity after heart transplantation. Transplantation 1995;60:1462-6. 5. Lindenfeld J, Miller GG, Shakar SF, et al. Drug therapy in the heart transplant recipient: part II: immunosuppressive drugs. Circulation 2004;110:3858-65. 6. Boyle JM, Moualla S, Arrigain S, et al. Risks and outcomes of acute kidney injury requiring dialysis after cardiac transplantation. Am J Kidney Dis 2006;48:787-96. 7. Chacko RC, Harper RG, Gotto J, Young J. Psychiatric interview and psychometric predictors of cardiac transplant survival. Am J Psychiatry 1996;153:1607-12. 8. Muirhead J, Meyerowitz BE, Leedham B, Eastburn TE, Merrill WH, Frist WH. Quality of life and coping in patients awaiting heart transplantation. J Heart Lung Transplant 1992;11:265-71; discussion: 271–2. 9. Singh TP, Gauvreau K, Bastardi HJ, Blume ED, Mayer JE. Socioeconomic position and graft failure in pediatric heart transplant recipients. Circ Heart Fail 2009;2:160-5.
10. Sirri L, Magelli C, Grandi S. Predictors of perceived social support in long-term survivors of cardiac transplant: the role of psychological distress, quality of life, demographic characteristics and clinical course. Psychol Health 2011;26:77-94. 11. Bohachick P, Taylor MV, Sereika S, Reeder S, Anton BB. Social support, personal control, and psychosocial recovery following heart transplantation. Clin Nurs Res 2002;11:34-51. 12. Bunzel B, Wollenek G. Heart transplantation: are there psychosocial predictors for clinical success of surgery? Thorac Cardiovasc Surg 1994;42:103-7. 13. Allen JG, Weiss ES, Arnaoutakis GJ, et al. The impact of race on survival after heart transplantation: an analysis of more than 20,000 patients. Ann Thorac Surg 2010;89:1956-63; discussion: 1963– 4. 14. Matas M, Staley D, Griffin W. A profile of the noncompliant patient: a thirty-month review of outpatient psychiatry referrals. Gen Hosp Psychiatry 1992;14:124-30. 15. Beck DE, Fennell RS, Yost RL, Robinson JD, Geary D, Richards GA. Evaluation of an educational program on compliance with medication regimens in pediatric patients with renal transplants. J Pediatr 1980; 96:1094-7. 16. Flurkey K, Stadecker M, Miller RA. Memory T lymphocyte hyporesponsiveness to non-cognate stimuli: a key factor in age-related immunodeficiency. Eur J Immunol 1992;22:931-5. 17. Pawelec G, Barnett Y, Forsey R, et al. T cells and aging, January 2002 update. Front Biosci 2002;7:d1056-183. 18. De Geest S, Dobbels F, Martin S, Willems K, Vanhaecke J. Clinical risk associated with appointment noncompliance in heart transplant recipients. Prog Transplant 2000;10:162-8. 19. Grady KL, Costanzo MR, Fisher S, Koch D. Preoperative obesity is associated with decreased survival after heart transplantation. J Heart Lung Transplant 1996;15:863-71. 20. Nagele H, Kalmar P, Rodiger W, Stubbe HM. Smoking after heart transplantation: an underestimated hazard? Eur J Cardiothorac Surg 1997;12:70-4. 21. Lawrence E, Barry RA, Brock RL, et al. The Relationship Quality Interview: evidence of reliability, convergent and divergent validity, and incremental utility. Psychol Assess 2011;23:44-63. 22. Kaba E, Thompson DR, Burnard P. Coping after heart transplantation: a descriptive study of heart transplant recipients’ methods of coping. J Adv Nurs 2000;32:930-6.