Journal of Cardiac Failure Vol. 21 No. 7 2015
Improved Outcomes for Women on the Heart Transplant Wait List in the Modern Era ALANNA A. MORRIS, MD,1,2 ROBERT T. COLE, MD,1 S. RAJA LASKAR, MD,1 ANDREAS KALOGEROPOULOS, MD, PhD,1 J. DAVID VEGA, MD,1,2 ANDREW SMITH, MD,1 AND JAVED BUTLER, MD, MPH3 Atlanta, Georgia; and Stony Brook, New York
ABSTRACT Background: Whether the routine use of continuous-flow left ventricular assist devices (LVAD) has affected gender differences in outcomes for patients listed for heart transplantation (HT) is unclear. Methods and Results: We identified 20,468 adults (25% women) listed as status 1A or 1B for HT from 2000 to 2014. Sex differences in removal from the wait list during the first 365 days due to death or deterioration was assessed with the use of Kaplan-Meier survival analysis. Patients were stratified according to listing before (era 1) or after (era 2) Food and Drug Administration approval of the Heartmate II LVAD on April 22, 2008. Freedom from death or deterioration on the wait list was higher for men than for women (70% vs 64%; P ! .001). After adjusting for risk factors, women had a higher risk of removal from the wait list at 365 days during both era 1 (hazard ratio [HR] 1.23, 95% confidence interval [CI] 1.10e1.36; P ! .001) and era 2 (HR 1.15, 95% CI 1.01e1.31; P 5 .029). Further adjustment for LVAD use eliminated the higher risk for women in era 2 (HR 1.14, 95% CI 0.99e1.29; P 5 .053) and not in era 1 (HR 1.22, 95% CI 1.10e1.36; P ! .001). Conclusions: The higher risk for death or deterioration in women waiting for HT has improved in the modern era. (J Cardiac Fail 2015;21:555e560) Key Words: Sex, heart transplant, ventricular assist device.
listing, declining from 102.2 to 12.9 deaths per 100 wait list years from 2001 to 2011. The demographic characteristics of wait list candidates has changed in the past decade, with data from the Organ Procurement and Transplantation Network (OPTN) showing that the proportion of women on the wait list has increased over time.2 Previous studies have reported conflicting data on whether women are at a higher risk for death or deterioration on the waiting list.3e5 Sex-based disparities with higher risk for women have been reported for periprocedural morbidity and mortality,6,7 as well as for outcomes after VAD implantation.8 The first-generation pulsatile LVADs were limited by their considerable bulk, and could not be implanted into patients with body surface areas !1.5 m2, making women with smaller body habitus ineligible for that therapy. With the United States Food and Drug Administration (FDA) approval of the first continuous-flow LVAD (Heartmate II; Thoratec, Pleasonton, California) in April 2008, improvements in sex-based differences in wait list outcomes might be expected, because these smaller and more durable LVADs could be implanted in women at the same rates as in men. In the present study, we sought to assess whether wait list outcomes among men and women listed for HT have been influenced by greater use of the continuous-flow LVAD.
Heart transplantation (HT) remains the best option for select patients with end-stage heart failure (HF). Over the past decade, changes to the allocation algorithm and improvements in left ventricular assist device (LVAD) technology have led to improved outcomes for patients on the HT waiting list,1,2 with an overall reduction in wait list mortality from 16.9 to 11.6 deaths per 100 wait list years from 2001 to 2011.2 Similarly, wait list mortality has improved among candidates with an implanted LVAD at the time of
From the 1Division of Cardiology Department of Medicine, Emory University School of Medicine, Atlanta, Georgia; 2Division of Cardiology Department of Medicine, Atlanta VA Medical Center, Atlanta, Georgia and 3Division of Cardiology Department of Medicine, Stony Brook University, Stony Brook, New York. Manuscript received September 22, 2014; revised manuscript received January 20, 2015; revised manuscript accepted March 9, 2015. Reprint requests: Alanna A. Morris, MD, 1462 Clifton Road Suite 528A, Atlanta, GA 30322. Tel: 404-727-4037; Fax: 404-712-0149. E-mail: [email protected] Funding Sources: NIH/NHLBI 1 U10 HL110302 (Butler) and NIH/NINR 5 P01 NR011587e03 (Butler). All decisions regarding this manuscript were made by a guest editor. See page 560 for disclosure information. 1071-9164/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cardfail.2015.03.009
555
556 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 Methods Study Population All subjects $18 years of age listed for HT from January 2000 to March 2014 were identified in the OPTN database, which includes deidentified data on all patients listed for HT in the United States. The Health Resources and Services Administration and the United States Department of Health and Human Services provide oversight to the activities of the OPTN contractor, the United Network of Organ Sharing (UNOS). When listed for HT, patients are designated as UNOS status 1A, 1B, or 2 based on their degree of hemodynamic compromise. UNOS status 1A includes patients requiring VAD, total artificial heart (TAH), extracorporeal membrane oxygenation (ECMO), intra-aortic balloon pump (IABP), mechanical ventilation, high-dose continuous intravenous inotropes, or an exemption for critical illness, such as ventricular tachycardia or complications with mechanical circulatory support. UNOS status 1B is the next highest status, and includes patients receiving continuous intravenous inotropes as well as stable VAD patients. Study End Point The primary end point was removal from the wait list by 365 days after listing due to death or deterioration. The 365-day time period was chosen because O92% of patients were removed from the wait list either for transplantation or for death or deterioration by this time. The determination that a patient has clinically deteriorated and is too sick for a transplant is made by the transplant center and is not based on any set of clinical criteria defined by UNOS. Once the decision is made by the transplant center to remove the patient from the wait list, the wait list candidate and their family are informed, and this information is entered into the UNOS wait list removal database as ‘‘Candidate condition deteriorated, too sick for transplant.’’ Statistical Analysis Data are presented as median (interquartile range [IQR]), or as n (%) of patients. Baseline characteristics were compared between sexes with the use of the c2 test for categorical and the MannWhitney test for continuous variables. Unadjusted survival rates were assessed by means of the Kaplan-Meier method. Multivariable Cox proportional hazards models were developed with the use of a forward selection procedure retaining variables significant at the 0.20 level based on a likelihood ratio to test for the association of female sex and the primary end point. The proportional hazards assumption was tested and verified for all risk factors by means of Schoenfeld residual correlation analysis. Two models were considered. Model 1 adjusted for the following variables: age, white race, ABO blood group, HF etiology, body mass index (BMI), estimated glomerular filtration rate (eGFR), albumin, diabetes, insurance (private, Medicaid, Medicare, other), ventilator status, inotrope use, ECMO, IABP, pulmonary capillary wedge pressure (PCWP), cardiac index (CI), volume of listing center, and presence of defibrillator. Model 2 adjusted for type of VAD (LVAD vs other VAD [RVAD, biventricular assist device (BiVAD), or TAH]). To test the effect of era on wait list outcomes, era was added to models 1 and 2 as a binary variable using January 2000 to April 21, 2008 (era 1) as the reference group and April 22, 2008, to March 2014 (era 2) as the comparator group. We also performed a subgroup analysis on those patients who were listed for HT after
April 21, 2008. Sex-era interaction terms were added to the main effect models to assess whether any improvement in wait list survival over time was modified by sex. Stratified multivariable models were developed to confirm significant sex-era interactions in the overall model. Data were analyzed with the use of SPSS statistical software version 22 (IBM, Armonk, New York).
Results Study Population
During the study period, 21,938 patients $18 years of age were listed in the United States as status 1A or 1B for HT. Of these, 1,470 patients were removed from the wait list for reasons other than the end points of interest (Fig. 1) and were excluded from the current analysis. Of the remaining 20,468 patients that formed the study cohort, 5,038 (25%) were women. Table 1 summarizes the baseline characteristics at the time of listing for HT stratified according to sex. Women were younger, more likely to have nonischemic HF etiology, and less likely to be white, overweight/obese, have diabetes, or normal renal function. Specific nonischemic HF etiologies that were more common in women included adriamycin-induced (4.6% vs 0.5%), peripartum (6.8% vs 0%), and myocarditis (1.4% vs 0.6%). Women were more likely to require mechanical ventilation, and circulatory support with intravenous inotropes or ECMO, and they were less likely to have an implantable cardioverterdefibrillator (ICD) or to be supported with an LVAD. The median number of days on the wait list was shorter for women than for men. Wait List Outcomes
During the overall study period, 3,542 candidates were removed from the wait list for either death (n 5 2,399) or being deemed to be too sick for transplantation (n 5 1,143). At 365 days, freedom from death or deterioration on the wait list was higher for men than for women 21,938 patients assessed for eligibility
1,470 patients were excluded because of the following removal reasons: 73 Refused transplant 287 Were transferred to another transplant center 575 Condition improved 19 Removed in error 19 Unable to contact 497 Other
20,468 patients included in primary analysis Fig. 1. Flow chart showing reasons for patient exclusion.
Sex and Wait List Outcomes Table 1. Baseline Characteristics of Men and Women Listed for Heart Transplantation From 2000 to 2014 Men (n 5 15,430) Age, y 55 White race 10,629 Etiology of heart failure Ischemic 6,979 Dilated 6,988 Restrictive 243 Hypertrophic 182 Valvular 269 Congenital 254 Earlier transplant 414 Other 101 Era of listing January 2000 to 7,492 April 21, 2008 April 22, 2008, to 7,938 March 2014 UNOS status at listing 1A 6,028 1B 9,402 Total days on waiting list 49 Center volume, no. of transplants O300 6,250 150e299 5,206 !150 3,974 Insurance Private 8,456 Medicaid 1,783 Medicare 4,218 Other public 528 Other 311 ABO blood group AB 738 A 5,968 B 2,171 O 6,553 Body mass index, kg/m2 Normal 4,695 Underweight 142 Overweight 5,958 Obese 4,635 eGFR, mL min1 1.73 m2 $60 8,237 30e59 6,149 !30 1,044 Albumin 3.6 Diabetes 4,317 PCWP, mm Hg 20 2.1 CI, L min1 m2 Inotropes 8,395 ECMO 145 IABP 1,323 Ventilator 816 ICD 10,264 Type of ventricular assist device LVAD 3,086 RVAD 25 TAH 87 BiVAD 417 Other (unspecified) 1,191
Women (n 5 5,038)
P Value
(46e62) (69)
52 (39e59) 3,023 (60)
!.001 !.001 !.001
(45) (45) (1) (1) (2) (2) (3) (1)
1,141 3,102 148 109 135 137 203 63
(49)
2,391 (48)
(51)
2,647 (52)
(39) (61) (15e142)
1,996 (40) 3,042 (60) 35 (11e114)
(40) (34) (26)
1,990 (40) 1,786 (35) 1,262 (25)
(55) (12) (28) (3) (2)
2,684 858 1,238 107 95
(54) (17) (25) (2) (2)
(5) (39) (14) (42)
254 1,876 723 2,185
(5) (37) (15) (43)
(30) (1) (39) (30)
2,087 165 1,496 1,290
(41) (3) (30) (26)
(53) (40) (7) (3.2,3.9) (28) (17,27) (1.7,2.5) (54) (0.9) (9) (5) (67)
2,425 2,144 469 3.6 1,110 20 2.1 2,847 92 408 356 2,914
(48) (43) (9) (3.1,3.9) (22) (16,25) (1.7,2.6) (57) (1.8) (8) (7) (59)
(23) (62) (3) (2) (2) (3) (4) (1)
.5 !.001 .08
!.001
.3
781 11 17 171 342
Morris et al
557
(Fig. 2). After adjusting for the variables in model 1, the risk of removal from the wait list for death or deterioration was significantly higher in women than in men (Table 2). When VAD use was added to the model, the risk associated with female sex remained elevated. Other variables associated with a higher risk of death or deterioration are presented in Table 2. Era Effect
.2
(23) (0.2) (0.6) (3.1) (8.7)
!.001
Overall, the number of candidates removed from the wait list for death or deterioration improved from 2,026 (20.5%) in era 1 to 1,516 (14.3%) in era 2 (P ! .001 for comparison). After adjusting for the variables in model 2, there was a 30% reduction (HR 0.70, 95% CI 0.64e0.75; P ! .001) in the overall risk of removal from the wait list for death or deterioration in transplant candidates listed during era 2 (Table 2). Significant improvements in freedom from death or deterioration on the wait list occurred in both men and women listed during era 2 compared with those listed during era 1 (Fig. 3). Despite these improvements, however, freedom from death or deterioration on the wait list at 365 days was higher for men than for women in both eras. Despite overall increases in LVAD use during era 2, women were less likely than men to have an LVAD implantation during era 2 (24% vs 31%; P ! .001) as well as during era 1 (8% vs 11%; P 5 .002). In patients listed during era 2, the risk of removal from the wait list for death or deterioration was higher for women than men (HR 1.15, 95% CI 1.01e1.31; P 5 .029) in model 1. However, this risk improved after adjustment for LVAD use (HR 1.14, 95% CI 0.99e1.29; P 5 .053) in model 2. When sex-era interaction terms were added to the main effect model 2
!.001
.09 !.001 !.001 .03 .009 !.001 .3 !.001 !.001 !.001
(18) (0.2) (0.4) (3.8) (7.7)
UNOS, United Network for Organ Sharing; eGFR, estimated glomerular filtration rate; PCWP, pulmonary capillary wedge pressure; CI, cardiac index; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; ICD, implantable cardioverter defibrillator; LVAD, left ventricular assist device; RVAD, right ventricular assist device; TAH, total artificial heart; BiVAD, biventricular assist device. Data are presented as median (interquartile range) or n (%).
Fig. 2. Unadjusted Kaplan-Meier estimates of the risk of removal from the wait list for death or deterioration at 365 days stratified by sex.
558 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 Table 2. Multivariable Predictors of Risk of Removal From the Wait List for Death or Deterioration at 365 Days (Cox Regression) Model 1 HR (95% CI) Female sex 1.18 (1.08e1.28)y Age, y 40e59 Reference 18e39 0.92 (0.83e1.03) $60 1.24 (1.14e1.34)y White Race 0.95 (0.88e1.03) Etiology of heart failure (HF) Ischemic Reference Dilated 0.88 (0.81e0.95)* Restrictive 1.81 (1.45e2.27)y Hypertrophic 1.08 (0.79e1.48) Valvular 1.35 (1.08e1.69)* Congenital 1.34 (1.04e1.72)* Earlier transplant 1.42 (1.01e1.99)* Other 1.96 (1.68e2.29)y Era of listing January 2000 to Reference April 21, 2008 April 22, 2008, 0.65 (0.61e0.71)y to March 2014 Center volume, no. of transplants O300 Reference 150e299 1.14 (1.05e1.24)* !150 1.28 (1.17e1.39)y Insurance Private Reference Medicaid 1.20 (1.08e1.34)y Medicare 1.15 (1.06e1.25)y Other public 1.18 (0.97e1.43) Other 1.23 (0.98e1.54) ABO blood group AB Reference A 0.92 (0.76e1.11) B 0.88 (0.71e1.08) O 0.94 (0.78e1.14) Body mass index Normal Reference Underweight 1.25 (0.96e1.63) Overweight or obese 0.94 (0.86e1.02) Obese 0.95 (0.87e1.04) eGFR, mL min1 1.73 m2 $60 Reference 30e59 1.57 (1.45e1.70)y #30 2.31 (2.07e2.58)y Albumin, g/dL 0.80 (0.76e0.84)y Diabetes 1.10 (1.01e1.19)* PCWP O15 mm Hg 1.47 (1.32e1.64)y Cardiac index 0.99 (0.94e1.04) Mechanical ventilation 2.06 (1.84e2.31)y Inotropes 1.10 (1.02e1.19)* ECMO 3.15 (2.59e3.83)y IABP 1.62 (1.46e1.79)y ICD 0.86 (0.80e0.93)y VAD d None LVAD Other VAD
Model 2 HR (95% CI) 1.17 (1.08e1.27)y Reference 0.92 (0.82e1.02) 1.24 (1.14e1.35)y 0.95 (0.88e1.03) Reference 0.88 (0.81e0.96)* 1.77 (1.41e2.21)y 1.05 (0.77e1.44) 1.33 (1.06e1.67)* 1.31 (1.02e1.68)* 1.36 (0.97e1.91) 1.89 (1.61e2.21)y Reference 0.70 (0.64e0.75)y Reference 1.13 (1.04e1.23)* 1.28 (1.17e1.39)y Reference 1.21 (1.08e1.35)y 1.15 (1.06e1.25)y 1.18 (0.97e1.43) 1.20 (0.96e1.51) Reference 0.93 (0.76e1.13) 0.89 (0.72e1.09) 0.95 (0.79e1.15) Reference 1.25 (0.96e1.63) 0.95 (0.87e1.03) 0.96 (0.88e1.06) Reference 1.55 (1.44e1.68)y 2.25 (2.02e2.52)y 0.81 (0.77e0.85)y 1.08 (1.00e1.17)* 1.43 (1.29e1.59)y 1.00 (0.94e1.05) 2.06 (1.83e2.32)y 1.02 (0.94e1.10) 3.05 (2.51e3.71)y 1.60 (1.44e1.77)y 0.87 (0.81e0.94)y Reference 0.67 (0.60e0.76)y 1.12 (1.01e1.25)*
Abbreviations as in Table 1. Model 1 adjusted for sex, age, race, ABO blood group, HF etiology, body mass index, eGFR, albumin, diabetes, insurance, ventilator status, inotrope use, ECMO, IABP, PCWP, CI, volume of listing center, and presence of ICD. Model 2 further adjusted for type of VAD (LVAD vs other [RVAD, BiVAD, TAH]). *P # .05. y P # .001.
Fig. 3. Unadjusted Kaplan-Meier estimates of the risk of removal from the wait list for death or deterioration at 365 days stratified by sex and era of listing.
in Table 2, they were not statistically significant (sex-era interaction: P 5 .6), confirming overall improvements in outcomes for men and women in the cohort largely due to era effect, and a similar pattern of higher risk in women that was attenuated by adjustment for LVAD therapy (Fig. 4). Discussion In the present study, we analyzed differences in wait list survival based on sex and use of continuous flow LVADs. There are 2 major findings from our analysis: 1) the risk of removal from the wait list for death or deterioration has improved substantially in men and women in the modern era since FDA approval of the Heartmate II LVAD; and 2) the higher risk of removal from the wait list for death or deterioration in women appears to be improving in the modern era, even after adjusting for differences in baseline risk factors. Our research findings of higher mortality in women on the wait list have been seen in earlier studies. The Waiting for a New Heart Study prospectively evaluated 318 patients (18% female) newly listed for HT at 17 hospitals in Germany and Austria.3 The authors found that women had worse survival while awaiting transplant, even after adjusting for confounders. Hsich et al examined wait list outcomes in 28,852 patients (24% women) listed for HT in the United States from 2000 to 2010.5 The cohort included status 1A and 1B patients in addition to status 2 patients. The authors found that female sex was associated with an increased risk of death among status 1A patients, but was protective from death among status 2 patients. Our study adds to this literature, confirming that sex-specific
Sex and Wait List Outcomes
Fig. 4. Hazard ratios (HRs) and 95% confidence intervals (CIs) for women’s risk of removal from the wait list for death or deterioration at 365 days. Men are the referent group. Squares: era 1; circles: era 2. Model 1 adjusted for sex, age, race, ABO blood group, heart failure etiology, body mass index, estimated glomerular filtration rate, albumin, diabetes, insurance, ventilator status, inotrope use, extracorporeal membrane oxygenation, intra-aortic balloon pump, pulmonary capillary wedge pressure, cardiac index, volume of listing center, and presence of implantable cardioverterdefibrillator. Model 2 further adjusted for left ventricular assist device versus none or other ventricular assist device (right ventricular assist device, biventricular assist device, total artificial heart).
disparities in wait list survival exist, but our analysis suggests that these disparities appear to be improving in the modern era. Sex-specific data on mechanical circulatory support are somewhat limited. Women were woefully underrepresented in the initial clinical trials of pulsatile-flow VADs, largely owing to the fact that the older-generation devices were larger and less accommodating in women with smaller body size.9e11 More recent data of patients with newergeneration continuous-flow LVADs have shown similar rates of survival in men and women. An analysis of 401 women (pulsatile devices 5 78) and 1,535 men (pulsatile devices 5 402) from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) showed no statistically significant difference in mortality for either pulsatile- or continuous-flow devices based on sex in adjusted and unadjusted models.12 A survival analysis of 465 patients who received the Heartmate II LVAD as bridge to transplantation showed no differences in survival between women and men while on LVAD support in the first 18 months.13 Adverse events were similar, with the exception of hemorrhagic stroke, which occurred more frequently in women (0.10 vs 0.04 events/patient-year; P 5 .02), and device-related infections, which occurred less frequently in women (0.23 vs 0.44 events/patient-year; P 5 .006). Our cohort also included patients with other types of VADs, including RVAD, BiVAD, and TAH. The overall rates of use of these devices were relatively low and appeared to be similar between men and women, but there are few existing data to compare our findings with. We did observe a higher use of ECMO in female patients in both eras. We
Morris et al
559
hypothesize that this may be related to higher rates of acute causes of nonischemic HF that are associated with rapid deterioration, as is commonly seen in peripartum cardiomyopathy or acute myocarditis, or it may be secondary to the decreased use of durable VADs in this population. Other clinical parameters, such as the higher prevalence of underweight and advanced renal dysfunction that we observed, might make women less suitable candidates for VAD therapy. Still, the similar rates of survival between men and women after implantation of durable LVADs seen in contemporary analyses should translate into more aggressive use of the newer-generation devices in women. Other risk factors for death or deterioration in our analysis included the etiology of HF, the level of ventilatory and circulatory support, clinical variables reflecting the severity of HF (including PCWP and impaired renal function), diabetes, lower center volume, and public insurance including Medicaid and Medicare. Risk factors not captured in the OPTN database, such as differences in access to care and rate of disease progression, also could have contributed to wait list outcomes. ICDs were noted to be protective against the risk of death or deterioration on the wait list. We did note a lower prevalence of ICDs in women in our cohort, and multiple previous studies have shown a lower rate of ICD implantation in women compared with men.14e17 Sex-specific HF etiologies, such as peripartum cardiomyopathy, might limit the use of ICDs owing to anticipation of myocardial recovery. Additionally, the patients in the present study were presumably more likely to die from progressive pump failure than from sudden cardiac death. Nevertheless, most of the patients in our cohort likely meet guideline-based criteria for the use of ICD therapy, so it is troubling that these sex-based disparities exist in such a high-risk population. There are multiple limitations of our analysis. Although the use of registry data allows access to large numbers of patients with ‘‘real-world’’ outcomes, we were unable to confirm center compliance with data entry, nor could we confirm events such as death/HT with external sources, such as the Social Security Death Index. By limiting our analysis to those patients who were initially listed as status 1A or 1B, we did not examine outcomes on patients listed as status 2, nor did we capture status changes, ie, patients who were downgraded from status 1A to status 1B or even 2. However, we thought that limiting our analysis to status 1 patients would capture patients with the most advanced HF, given that in the modern era, status 2 patients have a much lower mortality than status 1 patients.18 In addition, status 2 patients constitute a very small proportion of the patients who receive LVADs.19 We also lacked information on cause of death, because a higher proportion of noncardiac death in female candidates would not promote more aggressive use of device-based therapies in this population. This analysis also lacks information on sex-based differences in patient preferences for device-based therapies. An analysis of 91 patients with HF showed no sexbased differences in preference for treatment options
560 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 including inotropes, LVAD, and standard medical management.20 In the nonrandomized INTREPID trial of 55 inotrope-dependent patients who were ineligible for HT, there were more women in the optimal medical therapy (OMT) arm than in the LVAD arm (28% vs 8%; P 5 .1). A certain proportion of patients in the OMT group did not receive an LVAD based on their choice not to undergo LVAD implantation. However, the authors did not report the distribution of reasons patients were included in the OMT arm, rendering it difficult to know whether women were more likely to be represented in the OMT group based on preference or other reasons. In conclusion, wait list outcomes are improving in the modern era, including the still disparate outcomes for women. A portion of the overall higher risk associated with female sex appears to be mediated by lower use of LVADs in women compared with men. However, more aggressive use of these life-saving therapies in women appears to be contributing to the reduction in wait list disparities. Further research is needed to fully understand the differences in outcomes according to sex. Disclosures None. Acknowledgments The data were supplied by UNOS as the contractor for the OPTN. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the OPTN or the US government. References 1. Singh TP, Almond CS, Taylor DO, Graham DA. Decline in heart transplant wait list mortality in the United States following broader regional sharing of donor hearts. Circ Heart Fail 2012;5:249e58. 2. Colvin-Adams M, Smith JM, Heubner BM, Skeans MA, Edwards LB, Waller C, et al. OPTN/SRTR 2011 annual data report: heart. Am J Transplant 2013;13:119e48. 3. Weidner G, Zahn D, Mendell NR, Smits JMA, Deng MC, Zittermann A, et al. Patients’ sex and emotional support as predictors of death and clinical deterioration in the Waiting for a New Heart Study: results from the 1-year follow-up. Prog Transplant 2011;21:106e14. 4. Bove AA, Kashem A, Cross RC, Wald J, Furukawa S, Berman GO, et al. Factors affecting survival after heart transplantation: comparison of pre- and post-1999 listing protocols. J Heart Lung Transplant 2006; 25:42e7. 5. Hsich EM, Starling RC, Blackstone EH, Singh TP, Young JB, Gorodeski EZ, et al. Does the UNOS heart transplant allocation system favor men over women? JACC Heart Fail 2014;2:347e55.
6. Hogue CW, Barzilai B, Pieper KS, Coombs LP, DeLong ER, Kouchoukos NT, et al. Sex differences in neurological outcomes and mortality after cardiac surgery: a Society of Thoracic Surgery national database report. Circulation 2001;103:2133e7. 7. Vaccarino V, Abramson JL, Veledar E, Weintraub WS. Sex differences in hospital mortality after coronary artery bypass surgery: evidence for a higher mortality in younger women. Circulation 2002;105:1176e81. 8. Kwon MH, Moriguchi JD, Ardehali A, Jocson R, Marelli D, Laks H, et al. Use of ventricular assist device as a bridge to cardiac transplantation: impact of age and other determinants on outcomes. Tex Heart Inst J 2009;36:214e9. 9. Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001;345:1435e43. 10. Rogers JG, Butler J, Lansman SL, Gass A, Portner PM, Pasque MK, et al. Chronic mechanical circulatory support for inotrope-dependent heart failure patients who are not transplant candidates: results of the INTREPID trial. J Am Coll Cardiol 2007;50:741e7. 11. Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy P, Radovancevic B, et al. Multicenter clinical evaluation of the Heartmate vented electric left ventricular assist system in patients awaiting heart transplantation. J Thorac Cardiovasc Surg 2001;122:1186e95. 12. Hsich EM, Naftel DC, Myers SL, Gorodeski EZ, Grady KL, Schmuhl D, et al. Should women receive left ventricular assist device support? Findings from INTERMACS. Circ Heart Fail 2012;5: 234e40. 13. Bogaev RC, Pamboukian SV, Moore SA, Chen L, John R, Boyle AJ, et al. Comparison of outcomes in women versus men using a continuous-flow left ventricular assist device as a bridge to transplantation. J Heart Lung Transplant 2011;30:515e22. 14. Klein L, Grau-Sepulveda MV, Bonow RO, Hernandez AF, Williams MV, Bhatt DL, et al. Quality of care and outcomes in women hospitalized for heart failure. Circ Heart Fail 2011;4:589e98. 15. Curtis LH, Al-Khatib SM, Shea AM, Hammill BG, Hernandez AF, Schulman KA. Sex differences in the use of implantable cardioverter-defibrillators for primary and secondary prevention of sudden cardiac death. JAMA 2007;298:1517e24. 16. Al-Khatib SM, Hellkamp AS, Hernandez AF, Fonarow GC, Thomas KL, Al-Khalidi HR, et al, Get With the Guidelines Steering Committee and Hospitals. Trends in use of implantable cardioverterdefibrillator therapy among patients hospitalized for heart failure: have the previously observed sex and racial disparities changed over time? Circulation 2012;125:1094e101. 17. Yancy CW, Fonarow GC, Albert NM, Curtis AB, Stough WG, Gheorghiade M, et al. Influence of patient age and sex on delivery of guideline-recommended heart failure care in the outpatient cardiology practice setting: findings from IMPROVE HF. Am Heart J 2009; 157:754e762.e752. 18. Wever-Pinzon O, Drakos SG, Kfoury AG, Nativi JN, Gilbert EM, Everitt M, et al. Morbidity and mortality in heart transplant candidates supported with mechanical circulatory support: is reappraisal of the current United Network for Organ Sharing thoracic organ allocation policy justified? Circulation 2013;127:452e62. 19. Kirklin JK, Naftel DC, Kormos RL, Stevenson LW, Pagani FD, Miller MA, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant 2013;32:141e56. 20. MacIver J, Rao V, Delgado DH, Desai N, Ivanov J, Abbey S, et al. Choices: a study of preferences for end-of-life treatments in patients with advanced heart failure. J Heart Lung Transplant 2008;27: 1002e7.
Improved Outcomes for Women on the Heart Transplant Wait List in the Modern Era ALANNA A. MORRIS, MD,1,2 ROBERT T. COLE, MD,1 S. RAJA LASKAR, MD,1 ANDREAS KALOGEROPOULOS, MD, PhD,1 J. DAVID VEGA, MD,1,2 ANDREW SMITH, MD,1 AND JAVED BUTLER, MD, MPH3 Atlanta, Georgia; and Stony Brook, New York
ABSTRACT Background: Whether the routine use of continuous-flow left ventricular assist devices (LVAD) has affected gender differences in outcomes for patients listed for heart transplantation (HT) is unclear. Methods and Results: We identified 20,468 adults (25% women) listed as status 1A or 1B for HT from 2000 to 2014. Sex differences in removal from the wait list during the first 365 days due to death or deterioration was assessed with the use of Kaplan-Meier survival analysis. Patients were stratified according to listing before (era 1) or after (era 2) Food and Drug Administration approval of the Heartmate II LVAD on April 22, 2008. Freedom from death or deterioration on the wait list was higher for men than for women (70% vs 64%; P ! .001). After adjusting for risk factors, women had a higher risk of removal from the wait list at 365 days during both era 1 (hazard ratio [HR] 1.23, 95% confidence interval [CI] 1.10e1.36; P ! .001) and era 2 (HR 1.15, 95% CI 1.01e1.31; P 5 .029). Further adjustment for LVAD use eliminated the higher risk for women in era 2 (HR 1.14, 95% CI 0.99e1.29; P 5 .053) and not in era 1 (HR 1.22, 95% CI 1.10e1.36; P ! .001). Conclusions: The higher risk for death or deterioration in women waiting for HT has improved in the modern era. (J Cardiac Fail 2015;21:555e560) Key Words: Sex, heart transplant, ventricular assist device.
listing, declining from 102.2 to 12.9 deaths per 100 wait list years from 2001 to 2011. The demographic characteristics of wait list candidates has changed in the past decade, with data from the Organ Procurement and Transplantation Network (OPTN) showing that the proportion of women on the wait list has increased over time.2 Previous studies have reported conflicting data on whether women are at a higher risk for death or deterioration on the waiting list.3e5 Sex-based disparities with higher risk for women have been reported for periprocedural morbidity and mortality,6,7 as well as for outcomes after VAD implantation.8 The first-generation pulsatile LVADs were limited by their considerable bulk, and could not be implanted into patients with body surface areas !1.5 m2, making women with smaller body habitus ineligible for that therapy. With the United States Food and Drug Administration (FDA) approval of the first continuous-flow LVAD (Heartmate II; Thoratec, Pleasonton, California) in April 2008, improvements in sex-based differences in wait list outcomes might be expected, because these smaller and more durable LVADs could be implanted in women at the same rates as in men. In the present study, we sought to assess whether wait list outcomes among men and women listed for HT have been influenced by greater use of the continuous-flow LVAD.
Heart transplantation (HT) remains the best option for select patients with end-stage heart failure (HF). Over the past decade, changes to the allocation algorithm and improvements in left ventricular assist device (LVAD) technology have led to improved outcomes for patients on the HT waiting list,1,2 with an overall reduction in wait list mortality from 16.9 to 11.6 deaths per 100 wait list years from 2001 to 2011.2 Similarly, wait list mortality has improved among candidates with an implanted LVAD at the time of
From the 1Division of Cardiology Department of Medicine, Emory University School of Medicine, Atlanta, Georgia; 2Division of Cardiology Department of Medicine, Atlanta VA Medical Center, Atlanta, Georgia and 3Division of Cardiology Department of Medicine, Stony Brook University, Stony Brook, New York. Manuscript received September 22, 2014; revised manuscript received January 20, 2015; revised manuscript accepted March 9, 2015. Reprint requests: Alanna A. Morris, MD, 1462 Clifton Road Suite 528A, Atlanta, GA 30322. Tel: 404-727-4037; Fax: 404-712-0149. E-mail: [email protected] Funding Sources: NIH/NHLBI 1 U10 HL110302 (Butler) and NIH/NINR 5 P01 NR011587e03 (Butler). All decisions regarding this manuscript were made by a guest editor. See page 560 for disclosure information. 1071-9164/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cardfail.2015.03.009
555
556 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 Methods Study Population All subjects $18 years of age listed for HT from January 2000 to March 2014 were identified in the OPTN database, which includes deidentified data on all patients listed for HT in the United States. The Health Resources and Services Administration and the United States Department of Health and Human Services provide oversight to the activities of the OPTN contractor, the United Network of Organ Sharing (UNOS). When listed for HT, patients are designated as UNOS status 1A, 1B, or 2 based on their degree of hemodynamic compromise. UNOS status 1A includes patients requiring VAD, total artificial heart (TAH), extracorporeal membrane oxygenation (ECMO), intra-aortic balloon pump (IABP), mechanical ventilation, high-dose continuous intravenous inotropes, or an exemption for critical illness, such as ventricular tachycardia or complications with mechanical circulatory support. UNOS status 1B is the next highest status, and includes patients receiving continuous intravenous inotropes as well as stable VAD patients. Study End Point The primary end point was removal from the wait list by 365 days after listing due to death or deterioration. The 365-day time period was chosen because O92% of patients were removed from the wait list either for transplantation or for death or deterioration by this time. The determination that a patient has clinically deteriorated and is too sick for a transplant is made by the transplant center and is not based on any set of clinical criteria defined by UNOS. Once the decision is made by the transplant center to remove the patient from the wait list, the wait list candidate and their family are informed, and this information is entered into the UNOS wait list removal database as ‘‘Candidate condition deteriorated, too sick for transplant.’’ Statistical Analysis Data are presented as median (interquartile range [IQR]), or as n (%) of patients. Baseline characteristics were compared between sexes with the use of the c2 test for categorical and the MannWhitney test for continuous variables. Unadjusted survival rates were assessed by means of the Kaplan-Meier method. Multivariable Cox proportional hazards models were developed with the use of a forward selection procedure retaining variables significant at the 0.20 level based on a likelihood ratio to test for the association of female sex and the primary end point. The proportional hazards assumption was tested and verified for all risk factors by means of Schoenfeld residual correlation analysis. Two models were considered. Model 1 adjusted for the following variables: age, white race, ABO blood group, HF etiology, body mass index (BMI), estimated glomerular filtration rate (eGFR), albumin, diabetes, insurance (private, Medicaid, Medicare, other), ventilator status, inotrope use, ECMO, IABP, pulmonary capillary wedge pressure (PCWP), cardiac index (CI), volume of listing center, and presence of defibrillator. Model 2 adjusted for type of VAD (LVAD vs other VAD [RVAD, biventricular assist device (BiVAD), or TAH]). To test the effect of era on wait list outcomes, era was added to models 1 and 2 as a binary variable using January 2000 to April 21, 2008 (era 1) as the reference group and April 22, 2008, to March 2014 (era 2) as the comparator group. We also performed a subgroup analysis on those patients who were listed for HT after
April 21, 2008. Sex-era interaction terms were added to the main effect models to assess whether any improvement in wait list survival over time was modified by sex. Stratified multivariable models were developed to confirm significant sex-era interactions in the overall model. Data were analyzed with the use of SPSS statistical software version 22 (IBM, Armonk, New York).
Results Study Population
During the study period, 21,938 patients $18 years of age were listed in the United States as status 1A or 1B for HT. Of these, 1,470 patients were removed from the wait list for reasons other than the end points of interest (Fig. 1) and were excluded from the current analysis. Of the remaining 20,468 patients that formed the study cohort, 5,038 (25%) were women. Table 1 summarizes the baseline characteristics at the time of listing for HT stratified according to sex. Women were younger, more likely to have nonischemic HF etiology, and less likely to be white, overweight/obese, have diabetes, or normal renal function. Specific nonischemic HF etiologies that were more common in women included adriamycin-induced (4.6% vs 0.5%), peripartum (6.8% vs 0%), and myocarditis (1.4% vs 0.6%). Women were more likely to require mechanical ventilation, and circulatory support with intravenous inotropes or ECMO, and they were less likely to have an implantable cardioverterdefibrillator (ICD) or to be supported with an LVAD. The median number of days on the wait list was shorter for women than for men. Wait List Outcomes
During the overall study period, 3,542 candidates were removed from the wait list for either death (n 5 2,399) or being deemed to be too sick for transplantation (n 5 1,143). At 365 days, freedom from death or deterioration on the wait list was higher for men than for women 21,938 patients assessed for eligibility
1,470 patients were excluded because of the following removal reasons: 73 Refused transplant 287 Were transferred to another transplant center 575 Condition improved 19 Removed in error 19 Unable to contact 497 Other
20,468 patients included in primary analysis Fig. 1. Flow chart showing reasons for patient exclusion.
Sex and Wait List Outcomes Table 1. Baseline Characteristics of Men and Women Listed for Heart Transplantation From 2000 to 2014 Men (n 5 15,430) Age, y 55 White race 10,629 Etiology of heart failure Ischemic 6,979 Dilated 6,988 Restrictive 243 Hypertrophic 182 Valvular 269 Congenital 254 Earlier transplant 414 Other 101 Era of listing January 2000 to 7,492 April 21, 2008 April 22, 2008, to 7,938 March 2014 UNOS status at listing 1A 6,028 1B 9,402 Total days on waiting list 49 Center volume, no. of transplants O300 6,250 150e299 5,206 !150 3,974 Insurance Private 8,456 Medicaid 1,783 Medicare 4,218 Other public 528 Other 311 ABO blood group AB 738 A 5,968 B 2,171 O 6,553 Body mass index, kg/m2 Normal 4,695 Underweight 142 Overweight 5,958 Obese 4,635 eGFR, mL min1 1.73 m2 $60 8,237 30e59 6,149 !30 1,044 Albumin 3.6 Diabetes 4,317 PCWP, mm Hg 20 2.1 CI, L min1 m2 Inotropes 8,395 ECMO 145 IABP 1,323 Ventilator 816 ICD 10,264 Type of ventricular assist device LVAD 3,086 RVAD 25 TAH 87 BiVAD 417 Other (unspecified) 1,191
Women (n 5 5,038)
P Value
(46e62) (69)
52 (39e59) 3,023 (60)
!.001 !.001 !.001
(45) (45) (1) (1) (2) (2) (3) (1)
1,141 3,102 148 109 135 137 203 63
(49)
2,391 (48)
(51)
2,647 (52)
(39) (61) (15e142)
1,996 (40) 3,042 (60) 35 (11e114)
(40) (34) (26)
1,990 (40) 1,786 (35) 1,262 (25)
(55) (12) (28) (3) (2)
2,684 858 1,238 107 95
(54) (17) (25) (2) (2)
(5) (39) (14) (42)
254 1,876 723 2,185
(5) (37) (15) (43)
(30) (1) (39) (30)
2,087 165 1,496 1,290
(41) (3) (30) (26)
(53) (40) (7) (3.2,3.9) (28) (17,27) (1.7,2.5) (54) (0.9) (9) (5) (67)
2,425 2,144 469 3.6 1,110 20 2.1 2,847 92 408 356 2,914
(48) (43) (9) (3.1,3.9) (22) (16,25) (1.7,2.6) (57) (1.8) (8) (7) (59)
(23) (62) (3) (2) (2) (3) (4) (1)
.5 !.001 .08
!.001
.3
781 11 17 171 342
Morris et al
557
(Fig. 2). After adjusting for the variables in model 1, the risk of removal from the wait list for death or deterioration was significantly higher in women than in men (Table 2). When VAD use was added to the model, the risk associated with female sex remained elevated. Other variables associated with a higher risk of death or deterioration are presented in Table 2. Era Effect
.2
(23) (0.2) (0.6) (3.1) (8.7)
!.001
Overall, the number of candidates removed from the wait list for death or deterioration improved from 2,026 (20.5%) in era 1 to 1,516 (14.3%) in era 2 (P ! .001 for comparison). After adjusting for the variables in model 2, there was a 30% reduction (HR 0.70, 95% CI 0.64e0.75; P ! .001) in the overall risk of removal from the wait list for death or deterioration in transplant candidates listed during era 2 (Table 2). Significant improvements in freedom from death or deterioration on the wait list occurred in both men and women listed during era 2 compared with those listed during era 1 (Fig. 3). Despite these improvements, however, freedom from death or deterioration on the wait list at 365 days was higher for men than for women in both eras. Despite overall increases in LVAD use during era 2, women were less likely than men to have an LVAD implantation during era 2 (24% vs 31%; P ! .001) as well as during era 1 (8% vs 11%; P 5 .002). In patients listed during era 2, the risk of removal from the wait list for death or deterioration was higher for women than men (HR 1.15, 95% CI 1.01e1.31; P 5 .029) in model 1. However, this risk improved after adjustment for LVAD use (HR 1.14, 95% CI 0.99e1.29; P 5 .053) in model 2. When sex-era interaction terms were added to the main effect model 2
!.001
.09 !.001 !.001 .03 .009 !.001 .3 !.001 !.001 !.001
(18) (0.2) (0.4) (3.8) (7.7)
UNOS, United Network for Organ Sharing; eGFR, estimated glomerular filtration rate; PCWP, pulmonary capillary wedge pressure; CI, cardiac index; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; ICD, implantable cardioverter defibrillator; LVAD, left ventricular assist device; RVAD, right ventricular assist device; TAH, total artificial heart; BiVAD, biventricular assist device. Data are presented as median (interquartile range) or n (%).
Fig. 2. Unadjusted Kaplan-Meier estimates of the risk of removal from the wait list for death or deterioration at 365 days stratified by sex.
558 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 Table 2. Multivariable Predictors of Risk of Removal From the Wait List for Death or Deterioration at 365 Days (Cox Regression) Model 1 HR (95% CI) Female sex 1.18 (1.08e1.28)y Age, y 40e59 Reference 18e39 0.92 (0.83e1.03) $60 1.24 (1.14e1.34)y White Race 0.95 (0.88e1.03) Etiology of heart failure (HF) Ischemic Reference Dilated 0.88 (0.81e0.95)* Restrictive 1.81 (1.45e2.27)y Hypertrophic 1.08 (0.79e1.48) Valvular 1.35 (1.08e1.69)* Congenital 1.34 (1.04e1.72)* Earlier transplant 1.42 (1.01e1.99)* Other 1.96 (1.68e2.29)y Era of listing January 2000 to Reference April 21, 2008 April 22, 2008, 0.65 (0.61e0.71)y to March 2014 Center volume, no. of transplants O300 Reference 150e299 1.14 (1.05e1.24)* !150 1.28 (1.17e1.39)y Insurance Private Reference Medicaid 1.20 (1.08e1.34)y Medicare 1.15 (1.06e1.25)y Other public 1.18 (0.97e1.43) Other 1.23 (0.98e1.54) ABO blood group AB Reference A 0.92 (0.76e1.11) B 0.88 (0.71e1.08) O 0.94 (0.78e1.14) Body mass index Normal Reference Underweight 1.25 (0.96e1.63) Overweight or obese 0.94 (0.86e1.02) Obese 0.95 (0.87e1.04) eGFR, mL min1 1.73 m2 $60 Reference 30e59 1.57 (1.45e1.70)y #30 2.31 (2.07e2.58)y Albumin, g/dL 0.80 (0.76e0.84)y Diabetes 1.10 (1.01e1.19)* PCWP O15 mm Hg 1.47 (1.32e1.64)y Cardiac index 0.99 (0.94e1.04) Mechanical ventilation 2.06 (1.84e2.31)y Inotropes 1.10 (1.02e1.19)* ECMO 3.15 (2.59e3.83)y IABP 1.62 (1.46e1.79)y ICD 0.86 (0.80e0.93)y VAD d None LVAD Other VAD
Model 2 HR (95% CI) 1.17 (1.08e1.27)y Reference 0.92 (0.82e1.02) 1.24 (1.14e1.35)y 0.95 (0.88e1.03) Reference 0.88 (0.81e0.96)* 1.77 (1.41e2.21)y 1.05 (0.77e1.44) 1.33 (1.06e1.67)* 1.31 (1.02e1.68)* 1.36 (0.97e1.91) 1.89 (1.61e2.21)y Reference 0.70 (0.64e0.75)y Reference 1.13 (1.04e1.23)* 1.28 (1.17e1.39)y Reference 1.21 (1.08e1.35)y 1.15 (1.06e1.25)y 1.18 (0.97e1.43) 1.20 (0.96e1.51) Reference 0.93 (0.76e1.13) 0.89 (0.72e1.09) 0.95 (0.79e1.15) Reference 1.25 (0.96e1.63) 0.95 (0.87e1.03) 0.96 (0.88e1.06) Reference 1.55 (1.44e1.68)y 2.25 (2.02e2.52)y 0.81 (0.77e0.85)y 1.08 (1.00e1.17)* 1.43 (1.29e1.59)y 1.00 (0.94e1.05) 2.06 (1.83e2.32)y 1.02 (0.94e1.10) 3.05 (2.51e3.71)y 1.60 (1.44e1.77)y 0.87 (0.81e0.94)y Reference 0.67 (0.60e0.76)y 1.12 (1.01e1.25)*
Abbreviations as in Table 1. Model 1 adjusted for sex, age, race, ABO blood group, HF etiology, body mass index, eGFR, albumin, diabetes, insurance, ventilator status, inotrope use, ECMO, IABP, PCWP, CI, volume of listing center, and presence of ICD. Model 2 further adjusted for type of VAD (LVAD vs other [RVAD, BiVAD, TAH]). *P # .05. y P # .001.
Fig. 3. Unadjusted Kaplan-Meier estimates of the risk of removal from the wait list for death or deterioration at 365 days stratified by sex and era of listing.
in Table 2, they were not statistically significant (sex-era interaction: P 5 .6), confirming overall improvements in outcomes for men and women in the cohort largely due to era effect, and a similar pattern of higher risk in women that was attenuated by adjustment for LVAD therapy (Fig. 4). Discussion In the present study, we analyzed differences in wait list survival based on sex and use of continuous flow LVADs. There are 2 major findings from our analysis: 1) the risk of removal from the wait list for death or deterioration has improved substantially in men and women in the modern era since FDA approval of the Heartmate II LVAD; and 2) the higher risk of removal from the wait list for death or deterioration in women appears to be improving in the modern era, even after adjusting for differences in baseline risk factors. Our research findings of higher mortality in women on the wait list have been seen in earlier studies. The Waiting for a New Heart Study prospectively evaluated 318 patients (18% female) newly listed for HT at 17 hospitals in Germany and Austria.3 The authors found that women had worse survival while awaiting transplant, even after adjusting for confounders. Hsich et al examined wait list outcomes in 28,852 patients (24% women) listed for HT in the United States from 2000 to 2010.5 The cohort included status 1A and 1B patients in addition to status 2 patients. The authors found that female sex was associated with an increased risk of death among status 1A patients, but was protective from death among status 2 patients. Our study adds to this literature, confirming that sex-specific
Sex and Wait List Outcomes
Fig. 4. Hazard ratios (HRs) and 95% confidence intervals (CIs) for women’s risk of removal from the wait list for death or deterioration at 365 days. Men are the referent group. Squares: era 1; circles: era 2. Model 1 adjusted for sex, age, race, ABO blood group, heart failure etiology, body mass index, estimated glomerular filtration rate, albumin, diabetes, insurance, ventilator status, inotrope use, extracorporeal membrane oxygenation, intra-aortic balloon pump, pulmonary capillary wedge pressure, cardiac index, volume of listing center, and presence of implantable cardioverterdefibrillator. Model 2 further adjusted for left ventricular assist device versus none or other ventricular assist device (right ventricular assist device, biventricular assist device, total artificial heart).
disparities in wait list survival exist, but our analysis suggests that these disparities appear to be improving in the modern era. Sex-specific data on mechanical circulatory support are somewhat limited. Women were woefully underrepresented in the initial clinical trials of pulsatile-flow VADs, largely owing to the fact that the older-generation devices were larger and less accommodating in women with smaller body size.9e11 More recent data of patients with newergeneration continuous-flow LVADs have shown similar rates of survival in men and women. An analysis of 401 women (pulsatile devices 5 78) and 1,535 men (pulsatile devices 5 402) from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) showed no statistically significant difference in mortality for either pulsatile- or continuous-flow devices based on sex in adjusted and unadjusted models.12 A survival analysis of 465 patients who received the Heartmate II LVAD as bridge to transplantation showed no differences in survival between women and men while on LVAD support in the first 18 months.13 Adverse events were similar, with the exception of hemorrhagic stroke, which occurred more frequently in women (0.10 vs 0.04 events/patient-year; P 5 .02), and device-related infections, which occurred less frequently in women (0.23 vs 0.44 events/patient-year; P 5 .006). Our cohort also included patients with other types of VADs, including RVAD, BiVAD, and TAH. The overall rates of use of these devices were relatively low and appeared to be similar between men and women, but there are few existing data to compare our findings with. We did observe a higher use of ECMO in female patients in both eras. We
Morris et al
559
hypothesize that this may be related to higher rates of acute causes of nonischemic HF that are associated with rapid deterioration, as is commonly seen in peripartum cardiomyopathy or acute myocarditis, or it may be secondary to the decreased use of durable VADs in this population. Other clinical parameters, such as the higher prevalence of underweight and advanced renal dysfunction that we observed, might make women less suitable candidates for VAD therapy. Still, the similar rates of survival between men and women after implantation of durable LVADs seen in contemporary analyses should translate into more aggressive use of the newer-generation devices in women. Other risk factors for death or deterioration in our analysis included the etiology of HF, the level of ventilatory and circulatory support, clinical variables reflecting the severity of HF (including PCWP and impaired renal function), diabetes, lower center volume, and public insurance including Medicaid and Medicare. Risk factors not captured in the OPTN database, such as differences in access to care and rate of disease progression, also could have contributed to wait list outcomes. ICDs were noted to be protective against the risk of death or deterioration on the wait list. We did note a lower prevalence of ICDs in women in our cohort, and multiple previous studies have shown a lower rate of ICD implantation in women compared with men.14e17 Sex-specific HF etiologies, such as peripartum cardiomyopathy, might limit the use of ICDs owing to anticipation of myocardial recovery. Additionally, the patients in the present study were presumably more likely to die from progressive pump failure than from sudden cardiac death. Nevertheless, most of the patients in our cohort likely meet guideline-based criteria for the use of ICD therapy, so it is troubling that these sex-based disparities exist in such a high-risk population. There are multiple limitations of our analysis. Although the use of registry data allows access to large numbers of patients with ‘‘real-world’’ outcomes, we were unable to confirm center compliance with data entry, nor could we confirm events such as death/HT with external sources, such as the Social Security Death Index. By limiting our analysis to those patients who were initially listed as status 1A or 1B, we did not examine outcomes on patients listed as status 2, nor did we capture status changes, ie, patients who were downgraded from status 1A to status 1B or even 2. However, we thought that limiting our analysis to status 1 patients would capture patients with the most advanced HF, given that in the modern era, status 2 patients have a much lower mortality than status 1 patients.18 In addition, status 2 patients constitute a very small proportion of the patients who receive LVADs.19 We also lacked information on cause of death, because a higher proportion of noncardiac death in female candidates would not promote more aggressive use of device-based therapies in this population. This analysis also lacks information on sex-based differences in patient preferences for device-based therapies. An analysis of 91 patients with HF showed no sexbased differences in preference for treatment options
560 Journal of Cardiac Failure Vol. 21 No. 7 July 2015 including inotropes, LVAD, and standard medical management.20 In the nonrandomized INTREPID trial of 55 inotrope-dependent patients who were ineligible for HT, there were more women in the optimal medical therapy (OMT) arm than in the LVAD arm (28% vs 8%; P 5 .1). A certain proportion of patients in the OMT group did not receive an LVAD based on their choice not to undergo LVAD implantation. However, the authors did not report the distribution of reasons patients were included in the OMT arm, rendering it difficult to know whether women were more likely to be represented in the OMT group based on preference or other reasons. In conclusion, wait list outcomes are improving in the modern era, including the still disparate outcomes for women. A portion of the overall higher risk associated with female sex appears to be mediated by lower use of LVADs in women compared with men. However, more aggressive use of these life-saving therapies in women appears to be contributing to the reduction in wait list disparities. Further research is needed to fully understand the differences in outcomes according to sex. Disclosures None. Acknowledgments The data were supplied by UNOS as the contractor for the OPTN. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the OPTN or the US government. References 1. Singh TP, Almond CS, Taylor DO, Graham DA. Decline in heart transplant wait list mortality in the United States following broader regional sharing of donor hearts. Circ Heart Fail 2012;5:249e58. 2. Colvin-Adams M, Smith JM, Heubner BM, Skeans MA, Edwards LB, Waller C, et al. OPTN/SRTR 2011 annual data report: heart. Am J Transplant 2013;13:119e48. 3. Weidner G, Zahn D, Mendell NR, Smits JMA, Deng MC, Zittermann A, et al. Patients’ sex and emotional support as predictors of death and clinical deterioration in the Waiting for a New Heart Study: results from the 1-year follow-up. Prog Transplant 2011;21:106e14. 4. Bove AA, Kashem A, Cross RC, Wald J, Furukawa S, Berman GO, et al. Factors affecting survival after heart transplantation: comparison of pre- and post-1999 listing protocols. J Heart Lung Transplant 2006; 25:42e7. 5. Hsich EM, Starling RC, Blackstone EH, Singh TP, Young JB, Gorodeski EZ, et al. Does the UNOS heart transplant allocation system favor men over women? JACC Heart Fail 2014;2:347e55.
6. Hogue CW, Barzilai B, Pieper KS, Coombs LP, DeLong ER, Kouchoukos NT, et al. Sex differences in neurological outcomes and mortality after cardiac surgery: a Society of Thoracic Surgery national database report. Circulation 2001;103:2133e7. 7. Vaccarino V, Abramson JL, Veledar E, Weintraub WS. Sex differences in hospital mortality after coronary artery bypass surgery: evidence for a higher mortality in younger women. Circulation 2002;105:1176e81. 8. Kwon MH, Moriguchi JD, Ardehali A, Jocson R, Marelli D, Laks H, et al. Use of ventricular assist device as a bridge to cardiac transplantation: impact of age and other determinants on outcomes. Tex Heart Inst J 2009;36:214e9. 9. Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001;345:1435e43. 10. Rogers JG, Butler J, Lansman SL, Gass A, Portner PM, Pasque MK, et al. Chronic mechanical circulatory support for inotrope-dependent heart failure patients who are not transplant candidates: results of the INTREPID trial. J Am Coll Cardiol 2007;50:741e7. 11. Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy P, Radovancevic B, et al. Multicenter clinical evaluation of the Heartmate vented electric left ventricular assist system in patients awaiting heart transplantation. J Thorac Cardiovasc Surg 2001;122:1186e95. 12. Hsich EM, Naftel DC, Myers SL, Gorodeski EZ, Grady KL, Schmuhl D, et al. Should women receive left ventricular assist device support? Findings from INTERMACS. Circ Heart Fail 2012;5: 234e40. 13. Bogaev RC, Pamboukian SV, Moore SA, Chen L, John R, Boyle AJ, et al. Comparison of outcomes in women versus men using a continuous-flow left ventricular assist device as a bridge to transplantation. J Heart Lung Transplant 2011;30:515e22. 14. Klein L, Grau-Sepulveda MV, Bonow RO, Hernandez AF, Williams MV, Bhatt DL, et al. Quality of care and outcomes in women hospitalized for heart failure. Circ Heart Fail 2011;4:589e98. 15. Curtis LH, Al-Khatib SM, Shea AM, Hammill BG, Hernandez AF, Schulman KA. Sex differences in the use of implantable cardioverter-defibrillators for primary and secondary prevention of sudden cardiac death. JAMA 2007;298:1517e24. 16. Al-Khatib SM, Hellkamp AS, Hernandez AF, Fonarow GC, Thomas KL, Al-Khalidi HR, et al, Get With the Guidelines Steering Committee and Hospitals. Trends in use of implantable cardioverterdefibrillator therapy among patients hospitalized for heart failure: have the previously observed sex and racial disparities changed over time? Circulation 2012;125:1094e101. 17. Yancy CW, Fonarow GC, Albert NM, Curtis AB, Stough WG, Gheorghiade M, et al. Influence of patient age and sex on delivery of guideline-recommended heart failure care in the outpatient cardiology practice setting: findings from IMPROVE HF. Am Heart J 2009; 157:754e762.e752. 18. Wever-Pinzon O, Drakos SG, Kfoury AG, Nativi JN, Gilbert EM, Everitt M, et al. Morbidity and mortality in heart transplant candidates supported with mechanical circulatory support: is reappraisal of the current United Network for Organ Sharing thoracic organ allocation policy justified? Circulation 2013;127:452e62. 19. Kirklin JK, Naftel DC, Kormos RL, Stevenson LW, Pagani FD, Miller MA, et al. Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant 2013;32:141e56. 20. MacIver J, Rao V, Delgado DH, Desai N, Ivanov J, Abbey S, et al. Choices: a study of preferences for end-of-life treatments in patients with advanced heart failure. J Heart Lung Transplant 2008;27: 1002e7.
