http://www.jhltonline.org

1053-2498/$ - see fronhttp://dx.doi.org/10.10

E-mail address: kg

Reprint requests:School of Medicine,Bluhm CardiovascularSt, Galter Pavilion 11-4860. Fax: 312-695-19

Overall quality of life improves to similar levelsafter mechanical circulatory support regardless ofseverity of heart failure before implantation

Kathleen L. Grady, PhD, APN, FAAN,a David Naftel, PhD,b

Lynne Stevenson, MD,c Mary Amanda Dew, PhD,d Gerdi Weidner, PhD,e

Francis D. Pagani, MD, PhD,f James K. Kirklin, MD,b Susan Myers, BBA,b

Timothy Baldwin, PhD,g and James Young, MDh

From the aDivision of Cardiac Surgery, Department of Surgery, Northwestern University, Chicago, Illinois; the bDepartment of Surgery,

University of Alabama at Birmingham, Birmingham, Alabama; the cCardiovascular Division, Brigham and Women's Hospital, Boston,Massachusetts; the dDepartments of Epidemiology, Biostatistics, and Psychiatry, University of Pittsburgh School of Medicine and Medical Center, Pittsburgh, Pennsylvania; the eDepartment of Biology, San Francisco State University, San Francisco, California; the fCenter for Circulatory Support, University of Michigan, Ann Arbor, Michigan; gthe National Heart, Lung, and Blood Institute, Bethesda, Maryland; hCleveland Clinic Foundation, Cleveland, Ohio.

KEYWORDS:mechanical circulatorysupport;health-related qualityof life;heart failure;INTERMACS;continuous-flow pump

t matter r 2014 I16/j.healun.2013.1

rady@nmh.org

Kathleen L. GradNorthwestern UniInstitute, Division140, Chicago, IL03.

BACKGROUND: The severity of pre-implantation heart failure may affect post-implantation health-related quality of life (HRQOL). The purpose of our study was to examine differences in HRQOL frombefore mechanical circulatory support (MCS) through 1 year after surgery by Interagency Registry forMechanically Assisted Circulatory Support (INTERMACS) patient profiles.METHODS: Data from 1,559 adults with advanced heart failure who received primary continuous-flowpumps between June 23, 2006, and March 31, 2010, and were enrolled in INTERMACS were analyzed.HRQOL data were collected using the EQ-5D-3L survey before implantation and at 3, 6, and 12 monthsafter implantation. Statistical analyses included chi-square and t-tests, using all available data for eachtime period. Paired t-tests and sensitivity analyses were also conducted.RESULTS: HRQOL was poor before MCS implantation among patients with INTERMACS profiles1 to 7 and significantly improved after MCS implantation for all profiles. Stratified by INTERMACSprofile, problems within each of the 5 dimensions of HRQOL (i.e., mobility, self-care, usual activities,pain, and anxiety/depression) generally decreased from before to after implantation. By 6 monthsafter implantation, patients with all INTERMACS profiles reported similar frequencies of problemsfor all HRQOL dimensions. Paired t-tests and sensitivity analyses supported almost all of ourfindings.CONCLUSIONS: HRQOL is poor among advanced heart failure patients with INTERMACS profiles 1 to7 before MCS implantation and improves to similar levels for patients who remained on MCS 1 yearafter surgery. Patients have problems in HRQOL dimensions before and after MCS; however, thefrequency of reporting problems decreases for all dimensions within most profiles across time.J Heart Lung Transplant 2014;33:412–421r 2014 International Society for Heart and Lung Transplantation. All rights reserved.

nternational Society for Heart and Lung0.017

y, PhD, APN, FAAN, Feinbergversity, Center for Heart Failure,of Cardiac Surgery, 201 E Huron60611-3056. Telephone: 312-695-

The severity of heart failure has traditionally been charac-terized using the New York Heart Association (NYHA)functional classification system and gauges the severity ofsymptoms for patients who have American Heart AssociationStage C or D heart failure.1 Patients with advanced heart

Transplantation. All rights reserved.

Table 1 Severity of Advanced Heart Failure in Patients BeingConsidered for Mechanical Circulatory Support CharacterizedAccording to Interagency Registry for Mechanically AssistedCirculatory Support profiles2

Profile Description

1 Critical cardiogenic shock� Life-threatening hypotension and rapidly escalatinginotropic pressor support

2 Progressive decline� Dependent on inotropic support, with signs ofdeterioration

3 Stable but inotrope-dependent� Clinically stable on mild-moderate doses ofintravenous inotropes

4 Resting symptoms� At home on oral therapy, but frequently hassymptoms of congestion

5 Exertion intolerant� Comfortable at rest but unable to engage in anyactivity

6 Exertion limited� Comfortable at rest without evidence of fluidoverload, able to do mild activity

7 Advanced New York Heart Association class III� Clinically stable with a reasonable level ofcomfortable activity

Grady et al. QOL Improves After MCS 413

failure with NYHA class III or IV symptoms have limitationsin daily activities despite optimal medical management. How-ever, not all patients with advanced heart failure have asimilar prognosis. Investigators of the Interagency Registryfor Mechanically Assisted Circulatory Support (INTER-MACS), a prospective registry that collects data on allpatients who receive a durable, U.S. Food and DrugAdministration-approved mechanical circulatory support(MCS) device in the United States, developed a classificationsystem of patient profiles to further characterize the severityof heart failure in this high-risk population of patientsundergoing treatment with MCS.2 Seven patient profiles (1 ¼critical cardiogenic shock to 7 ¼ advanced NYHA class IIIsymptoms) were defined by clinical descriptors at the time ofMCS implantation.2,3

INTERMACS profiles have been used to characterizerisk for poor outcomes after MCS implantation. Stepanenkoet al,4 in their study on outcomes of elective vs emergentpermanent (i.e., destination therapy) MCS in elderly patientswith advanced heart failure, reported that patients notrequiring inotrope support (INTERMACS profiles 4–7) hadlower operative mortality than patients with cardiogenicshock on high-dose inotropes (INTERMACS profiles 1–3),cautiously concluding that permanent MCS may benefithighly selected elderly patients. In another study, patientswith INTERMACS profiles 4–7 had better actuarial survivaland a shorter hospital length of stay than patients withINTERMACS profile 1.5 Kirklin et al6 reported thatINTERMACS profile 1 is a risk factor for death early afterMCS implantation as destination therapy.

Researchers have demonstrated improved patient-reported health-related quality of life (HRQOL) outcomesthrough as long as 2 years after MCS implantation forbridge to transplant and destination therapy.7–10 However,improvement in HRQOL from before to after MCSimplantation across the spectrum of heart failure severityis unknown. Pre-implantation heart failure severity affectspost-implantation clinical outcomes and thus may affectpost-implantation HRQOL outcomes as well. DeterminingMCS HRQOL benefits and risks, stratified by heart failureseverity, may provide important information with which toinform patients who are considering MCS implantation andthereby help patients make more informed health caredecisions, given that severely ill patients make decisions ontreatment options (e.g., surgical therapies or continuedmedical therapy, palliative care, or hospice) based onoutcomes that include HRQOL as well as survival.11–14

Such information may also guide development of moretargeted strategies to enhance HRQOL after implantation.Understanding the HRQOL benefits of MCS is especiallyimportant for patients who receive MCS as destinationtherapy. In the future, indications for MCS as destinationtherapy may expand to include patients with moderatelyadvanced heart failure, thus providing further impetus forunderstanding variation in post-MCS HRQOL outcomes bypre-implantation advanced heart failure severity.15

The purpose of our study was to examine differences inHRQOL outcomes from before MCS implantation through1 year after surgery among patients receiving continuous-

flow pumps, stratified by INTERMACS patient profiles. Wedefined HRQOL as “the functional effect of an illness andits consequent therapy upon a patient, as perceived by thepatient.”16

Methods

Sample

The source of data for this study was INTERMACS. Studyinclusion criteria were adult patients (age Z19 years at time ofimplantation) who received a primary implant between June 23,2006, and March 31, 2010, and whose left ventricular assist devicewas a continuous-flow pump. Excluded were 9 patients from4 hospitals that did not collect post-implantation HRQOL data. Thefollow-up date for this study was March 31, 2011, allowing theopportunity for each patient to have 1 year of post-implantationfollow-up. These criteria resulted in a sample of 1,559 patientsfrom 101 hospitals. Patients were categorized according toINTERMACS profiles (Table 1).

Instruments

HRQOL was measured using the EQ-5D-3L survey, a generic,self-report HRQOL instrument that includes 5 questions measuringthe HRQOL dimensions of mobility, self-care, usual activities,pain/discomfort, and anxiety/depression.17,18 For each dimension,patients choose among 3 levels of Likert-scaled responses: noproblems, some or moderate problems, and extreme problems. TheEQ-5D also has a newer 5-level response version for the 5 dimensions(EQ-5D-5L) that is more precise and has been available since April

The Journal of Heart and Lung Transplantation, Vol 33, No 4, April 2014414

2011. Because we have 5 years of INTERMACS data using the3-level response version and determined that it is not appropriate tomap from a 3-level response format to a 5-level response format, wecontinued to use the EQ-5D-3L. The EQ-5D-3L also includes anoverall health status rating, using a vertical visual analog scale(VAS), with 0 ¼ worst imaginable health state and 100 ¼ bestimaginable health state. Psychometric support for this instrument hasbeen reported.19 Also collected were data from medical records,including demographic and behavioral characteristics and pre-implantation clinical characteristics.

Procedures

All sites participating in INTERMACS received InstitutionalReview Board approval. Patients provided consent and weresubsequently enrolled in the registry. For this report, EQ-5D-3Ldata and medical records data were collected by researchcoordinators before and at 3, 6, and 12 months after implantationuntil device removal, transplant, or death during this 12-monthfollow-up period. Patients completed the EQ-5D as close to theimplantation date as possible, up to 30 days prior. Reasons formissing data were recorded when the EQ-5D-3L was notcompleted. Data were entered electronically into the INTERMACSdatabase and analyzed by the data coordinating center, located atthe University of Alabama, Birmingham, Alabama.

Statistical analyses

Patients with INTERMACS profiles 5, 6, and 7 were combinedinto 1 group due to the small sample sizes in each group andsufficient similarity in severity of illness (i.e., being exertionintolerant, exertion limited, or advanced NYHA class III). Datawere analyzed using SAS 9.1 software (SAS Institute, Carey, NC).The EQ-5D-3L VAS score was reported as a mean � standarddeviation, and dimension scores were reported as frequencies.Analyses for dimensions were dichotomized (no problems vsmoderate or extreme problems), and frequencies of moderate andextreme problems were displayed graphically.

Statistical analyses included chi-square tests to compareproportions and t-tests to compare means, using all available datafor each time period. Paired t-tests were used to test the robustnessof our analyses for a sub-set of patients with complete pre-implantation and 12-month post-implantation HRQOL data. Levelof significance was po 0.05 for longitudinal comparisons betweenpre-implantation and 12-month post-implantation time periods, aswell as for cross-sectional comparisons for all profiles within agiven time period.

Data from the 5 dimensions were clustered into two groups:group 1—physical function/activities of daily living (mobility, self-care, and usual activities) and group 2—pain/emotions (pain/discomfort and anxiety/depression). This grouping was usedbecause a response level of “extreme problems” was assigned topatients who were too sick to respond (as reported by the patient orresearch coordinator) for the physical function/activities of dailyliving group. This response level was assigned post hoc to reducethe potential for overestimation of HRQOL in patients who werethe most severely ill. No assignment of responses was made for toosick patients for the pain/emotions group, because being too sickdoes not necessarily indicate extreme problems regarding pain ornegative emotions. A VAS rating of 0 was assigned to patients whoalso were too sick to respond. We performed sensitivity analyses toassess the influence of assignment of scores for patients too sick torespond.

Results

Description of cohort

There were 1,559 advanced HF patients enrolled inINTERMACS pre-implantation between 2006 and 2010who received a primary continuous-flow pump. All patientsincluded in this report had pre-implantation EQ-5D-3L data(or a value assigned if too sick to respond) and post-implantation EQ-5D data at one or more time periods. Pre-implantation, group 1 (i.e., the physical function/activities ofdaily living group) included 1,190 of the 1,559 patients(76.3%). Most group 1 patients were white, middle-agedmen who were married and well educated (Table 2). Beforeimplantation, 84% of these patients were in INTERMACSprofiles 1 to 3, and only 5% were in INTERMACS profiles5 to 7. The 822 pre-implantation group 2 patients (i.e., thepain/emotions group) had similar demographic and clinicalcharacteristics as group 1 (data not displayed); however,there were fewer INTERMACS profile 1 patients.

Differences in pre-implantation clinical characteristicswere detected for the 1,190 group 1 patients with EQ-5D-3Ldata vs the 369 patients without EQ-5D-3L data who werenot included in our study (Table 2). More group 1 patientswere NYHA class IV and on inotrope therapy and aventilator before implantation than patients without EQ-5D-3L data. Demographic characteristics, behavioral variables,and comorbidities were similar between the groups.

EQ-5D-3L survey completion rates overall and byINTERMACS patient profile

Rates of survey completion were examined for group 1 andgroup 2 patients by INTERMACS patient profiles. Data aredisplayed for group 1 patients in Table 3; data are not displayedfor group 2 patients. Differences in rates of survey completionwere detected for group 1 and group 2 patients across time for allINTERMACS profiles, except for INTERMACS profiles 5 to 7.Rates of survey completion also differed by INTERMACSprofiles within the pre-implantation and 6-month post-implanta-tion periods for group 1 and 2 patients. Common reasons for lackof survey completion were administrative (e.g., patient notconsented, no contact with the patient during the window of timethat a survey was due) and patient refusal to participate.

Differences in HRQOL among INTERMACS profilesbefore and after MCS implantation

Differences in HRQOL were examined by INTERMACSprofile from baseline through 12 months after implantation.Mean VAS scores before implantation revealed poor healthstatus among patients with all INTERMACS profiles,although scores differed significantly by profile (Figure 1).Patients with all INTERMACS profiles experienced sig-nificantly improved health status from before to 12 monthsafter implantation. Health status differed significantlyamong profiles at 3 months after implantation but wassimilar for all profiles at 6 and 12 months after implantation.

Table 2 Characteristics of Pre-implantation Mechanical Circulatory Support Recipients With and Without EQ-5D-3L Data

Pre-implantcharacteristics

Pre-implant MCS recipients

p-valueWith EQ-5D data Without EQ-5D data(n ¼ 1,190)a (n ¼ 369)

Demographic and behavioral characteristicsAge at implant, mean years 53.48 � 12.40 54.11 � 12.38 0.39Male, % 77.2 80.5 0.19Race, % white 69.4 65.9 0.20Married at time of implantation, % 62.3 62.3 1.004High school education, % 52.6 53.5 0.79Currently smoking, % 13.2 12.0 0.59Current alcohol abuse, % 15.2 18.3 0.16Current drug abuse, % 2.5 3.0 0.66

Clinical characteristics, %Primary cardiac diagnosisIschemic cardiomyopathy 36.1 35.0 0.68Dilated cardiomyopathy 49.7 56.8 0.04Other 14.2 9.2 0.01

Comorbidities, %Diabetes 37.9 40.1 0.46Cerebrovascular accident 8.1 5.3 0.08Right heart failure (RVEF severe) 24.3 26.3 0.60Pre-COPD 12.6 14.8 0.30NYHA class IV 79.6 71.1 0.0014Intra-aortic balloon pump 36.0 31.2 0.09Ventilator 9.7 4.1 0.0007ECMO 2.4 1.4 0.21Dialysis 2.1 2.7 0.49

INTERMACS profile at implant, %1 18.5 11.4 0.00142 45.9 40.4 0.063 19.8 25.5 0.024 10.7 13.0 0.215 2.4 4.6 0.036 1.7 3.0 0.127 1.0 2.2 0.0002

Device strategy, %Bridge to transplant

Listed 46.8 48.0 0.70Likely to be listed 30.5 33.1 0.35Moderately likely to be listed 9.9 9.2 0.69Unlikely to be listed 3.4 2.2 0.25

Destination therapy, % 7.9 6.5 0.38Inotrope therapy 83.6 74.8 0.0002Implantable cardioverter defibrillator 78.7 83.0 0.07Temporary circulatory support 12.9 11.6 0.56

COPD, chronic obstructive pulmonary disease; ECMO, extracorporeal membrane oxygenation; INTERMACS, Interagency Registry for Mechanically AssistedCirculatory Support; MCS, mechanical circulatory support; NYHA, New York Heart Association; RVEF, right ventricular ejection fraction.

aThose patients who were “too sick’ have been included in the “completed” column and assigned visual analog scale score of 0 and physical dimensionsas “extreme problems” (i.e., group 1).

Grady et al. QOL Improves After MCS 415

Physical function /activities of daily livingdimensions (group 1)

Stratified by INTERMACS profile, patient problems withinthe dimensions of HRQOL generally decreased from beforeto after implantation. Most patients with all INTERMACSprofiles had problems with mobility (Z 70%) and usualactivities (Z 80%) before implantation, whereas signifi-cantly fewer patients with all profiles had problems 12

months after implantation (Figure 2A and B). Although thefrequency of problems with self-care before implantationvaried significantly among profiles, significantly fewerpatients with all profiles had problems by 12 months afterimplantation, except profiles 5 to 7 (Figure 2C).

Differences in frequency of reporting problems byINTERMACS profile were also examined cross-sectionallyfor group 1 patients within each time period. Before implanta-tion, significant differences in problems with mobility, usual

Table 3 Primary Continuous-Flow Left Ventricular Assist Device Adults Who Completed the EQ-5D-3L by Patient Profile at Pre-implantand at 3 months, 6 months, and 12 Months Post-implanta

Patient profile

Pre-implant

Post-implant

p-value

3 months 6 months 12 months

No. QOLb % No. QOLb % No. QOLb % No. QOL* %

Level 1 262 220 84.0 235 113 48.1 211 73 34.6 141 49 34.8 o0.0001Level 2 695 546 78.6 649 323 49.8 590 251 42.5 425 172 40.5 o0.0001Level 3 330 236 71.5 322 149 46.3 301 133 44.2 228 96 42.1 o0.0001Level 4 175 127 72.6 163 78 47.9 155 67 43.2 110 54 49.1 o0.0001Levels 5–7 97 61 62.9 91 44 48.4 83 45 54.2 66 30 45.5 0.1Total 1,559 1,190 76.3 1,460 707 48.4 1,340 569 42.5 970 401 41.3p-value o0.0001 0.89 0.03 0.21

QOL, quality of life.aThe Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) patient profile level is collected before implant. This Table reports

the breakdown of patient profile level for those patients who completed the EQ-5D questionnaire at the specified time points. The p-values in the lastcolumn reflect differences in rates of completion for each INTERMACS profile across time. The p-values in the last row reflect differences in rates ofcompletion for each INTERMACS profile at a specified time period.

bPatients who answered the EQ-5D-3L. Patients who did not complete the EQ-5D because they were “too sick” were assigned the visual analog scalescore of 0 and the physical dimensions as extreme problems and were included in this table as EQ-5D form completions.

The Journal of Heart and Lung Transplantation, Vol 33, No 4, April 2014416

activities, and self-care were detected among the profiles(Figure 2A–C). At 3, 6, and 12 months after implantation, thefrequency of problems were generally similar for all profiles.

The presence of extreme vs some/moderate problems for theHRQOL dimensions is displayed graphically in Figure 2A–C.Although no formal analyses were conducted, empirically, morepatients with INTERMACS profile 1 had extreme problems vssome problems regarding mobility, usual activities, and self-carethan patients with other profiles before implantation.

Pain/emotion dimensions (group 2)

More than 50% of patients with all INTERMACS profilesreported having problems with pain before implantation,and significantly fewer patients reported pain at 12 months

Figure 1 EQ-5D visual analog scale (VAS) results according to accorSupport (INTERMACS) profiles. The range bars indicate the standard e

after implantation, except for patients with INTERMACSprofiles 1 and 4 (Figure 3A). Most patients with allINTERMACS profiles reported anxiety/depression beforeimplantation, which decreased significantly at 12 monthsafter implantation, except for profile 3, wherein there was astrong trend (Figure 3B).

Differences in frequency of reporting problems byINTERMACS profile were also examined for group2 patients cross-sectionally within each time period. Therewere almost no differences in frequency of reporting of painand anxiety/depression before implantation and at 3, 6, and12 months after implantation (Figure 3A–B).

Regarding empirically observed differences in levels ofproblems by INTERMACS profile for group 2 patients,most patients reported moderate problems, rather than extreme

ding to Interagency Registry for Mechanically Assisted Circulatoryrror.

Figure 2 Physical function/activities of daily living dimensions on the EQ-5D according to Interagency Registry for MechanicallyAssisted Circulatory Support (INTERMACS) profiles: (A) mobility problems; (B) unusual activities problems; and (C) self-care problems.The range bars indicate the standard error.

Grady et al. QOL Improves After MCS 417

Figure 3 Pain/emotions dimensions on the EQ-5D according to Interagency Registry for Mechanically Assisted Circulatory Support(INTERMACS) profiles: (A) pain/discomfort problems and (B) anxiety/depression problems. The range bars indicate the standard error.

The Journal of Heart and Lung Transplantation, Vol 33, No 4, April 2014418

problems, for pain and anxiety/depression before and afterimplantation (Figure 3A and B). Notably, almost 50% ofpatients with INTERMACS profiles 5 to 7 reported extremeproblems with anxiety/depression before implantation.

Comparisons for MCS patients with complete data

Comparisons using paired data for 348 patients before and at 12months after implantation supported almost all of our findings.

We found a similar pattern of significant increase in overallhealth status and decrease in problems for all 5 dimensions foralmost all profiles across time (Figure 4A–F). These pairedcomparisons did not reveal significant decreases in problemsfrom before to 12 months after implantation for INTERMACSprofiles 4 and 5 to 7 regarding self-care (Figure 4D), forINTERMACS profiles 3, 4, and 5 to 7 regarding pain(Figure 4E), and for INTERMACS profiles 3 and 4 regardinganxiety/depression (Figure 4F).

Figure 4 Comparisons for mechanical circulatory support patients with complete data on the EQ-5D according to Interagency Registryfor Mechanically Assisted Circulatory Support (INTERMACS) profiles: (A) visual analog scale; (B) mobility, any problems; (C) usualactivities, any problems; (D) self-care, any problems; (E) pain, any problems; (F) anxiety/depression, any problems. The range bars indicatethe standard error.

Grady et al. QOL Improves After MCS 419

Sensitivity analyses

Sensitivity analyses, without prospective assignment ofmissing data for patients too sick to respond, resulted insimilar findings for the VAS and the 3 dimensions ofmobility, usual activities, and self-care, thus supporting ourfindings.

Discussion

Our findings provide important new information regardingadvanced heart failure patients with the highest levels ofdisease severity and worst prognoses among patients withcardiovascular disease. Advanced heart failure patients withall INTERMACS profiles had poor overall HRQOL atbaseline, although HRQOL scores differed among profiles.At 1 year after surgery, HRQOL levels had improved, andpatients with all profiles had similar levels of HRQOL.Frequency of problems for all 5 dimensions of HRQOLdecreased significantly from before to 1 year after implan-tation for patients with all INTERMACS profiles, with fewexceptions. Similar to overall HRQOL, frequencies ofproblems did not differ by INTERMACS profile for all5 HRQOL dimensions at 12 months after implantation.

Almost all of our findings were supported by pairedt-tests and use of all available data. Specifically, using bothmethods of analyses, we found significant improvement inhealth status and a decrease in problems with mobility and

usual activities for all profiles, a decrease in problems withself-care for profiles 1, 2, and 3; a decrease in problems withpain for profile 2; and a decrease in problems with anxiety/depression for profiles 1, 2, and 5 to 7. Also, a lack of signi-ficance for decreases in problems was found using pairedt-tests as well as all available data for self-care (profiles 5–7),pain (profiles 1 and 4), and anxiety/depression (profiles 3 and4). Significant differences were detected when all availabledata were used, but no significant differences were detectedusing paired t-tests only for decreases in problems for self-care (profile 4) and pain (profiles 3 and 5–7). Thus, pairedt-tests supported the robustness of our analyses using allavailable data.

Our report examined HRQOL by heart failure severity frombefore to after MCS implantation. Prior reports have describedoverall improvement in HRQOL from before to early and laterafter implantation for first-generation and second-generationdevices.8,9,20–22 HRQOL outcomes by pre-operative heartfailure severity have also been examined in patients after hearttransplantation.23 We previously reported that patients, regard-less of heart failure severity before transplantation, were quitesatisfied with their overall HRQOL at 6 months after transplan-tation and reported similar and low levels of disability regardingmobility, self-care, and/or emotional status. Our findings in anMCS cohort are similar to these heart transplant patients.

However, MCS patients, no matter what profile, haveworse overall HRQOL before and after implantationcompared with a normative U.S. adult (4 18 years) general

The Journal of Heart and Lung Transplantation, Vol 33, No 4, April 2014420

population (U.S. adult mean VAS score is 82).24 Further-more, most of the U.S. adult population frequently report noproblems in the 5 EQ-5D dimensions (range, 74%–96%).25

Our findings, when examined within the context of the U.S.adult population, reveal the profound effect of advancedheart failure on HRQOL and yet also demonstrate improve-ment in HRQOL for all profiles after MCS.

Studies of HRQOL that have used the EQ-5D in otherpopulations with chronic illness provide additional insightinto our findings. Calvert et al26 reported low EQ-5D VASscores in primarily NYHA class III heart failure patients(range of mean VAS scores, stratified by age, 45–55). Thesemean VAS scores were not as low as the mean scores in ourcohort, even for patients at higher INTERMACS profiles.Almenar-Pertejo et al27 reported a pre-heart transplant meanEQ-5D VAS score of 37, and a mean VAS score range of 66to 71 through 1 year after surgery, similar to mean scores forpatients in our study. Patients from their study also reportedsignificant improvement in all HRQOL dimensions, exceptfor pain, from before to after heart transplantation.

Our findings demonstrate that regardless of disease seve-rity, HRQOL improves from before to 1 year after MCSimplantation. Understanding specific problems reported bypatients, no matter what disease severity, provides guidancefor clinicians to monitor potential areas of concern anddevelop interventions to enhance HRQOL before and afterMCS. For example, 4 30% of all patients reported pain/discomfort and anxiety/depression from baseline through1 year after MCS. Although fewer patients with someINTERMACS profiles reported problems in these dimen-sion after MCS, patients with other profiles reported similarlevels of problems before and after MCS. These findingsrequire further exploration with a more sensitive “problem-specific” instrument.

Regarding pain, it would be helpful to assess location ofpain, extent of pain, degree of relief from pain withtreatment, and the interference of pain with activities ofdaily living, sleep, etc. Regarding location of pain, giventhat post-MCS EQ-5D surveys are completed no soonerthan 3 months after implantation, we would not anticipatereports of pain related to a sternotomy or driveline tunnelingbut perhaps related to an infection at the driveline exit site.

Similarly, anxiety and depression may be mild, moder-ate, or severe, and further assessment with an instrumentspecific to each diagnosis would be useful to guide treat-ment options. Also, it would be important to understand anindividual patient’s clinical trajectory (e.g., before, early, orlater after MCS) and implant strategy regarding psycho-logic problems. For example, patients may be anxious aboutlearning self-care early after MCS, whereas other patientsmay become depressed if they experience a major com-plication later after MCS. Furthermore, bridge-to-transplantpatients may experience anxiety and/or depression regardinguncertainty about the timing of transplantation, whereasdestination therapy patients may experience anxiety and/ordepression about their length of life on MCS.

Our findings also provide insight into the HRQOL ofpatients with less severe heart failure. Given that patientswith INTERMACS profiles 5 to 7 have poor overall health

status and problems with mobility, usual activities, andanxiety, and/or depression pre-implantation, which improvesignificantly 12 months after implantation, our findingssupport investigation of HRQOL outcomes in patients withmoderately advanced heart failure across time.15

Our study has limitations. HRQOL data were collectedusing a brief generic health profile. An HRQOL instrumentspecific for heart failure might have provided additionalinformation. However, it is also important to note that in arecent review article, Dyer et al28 reported adequatepsychometric support (including validity, reliability, andresponsiveness) for use of the EQ-5D as an outcomemeasure within the cardiovascular area, including heartfailure. Our findings might also have been influenced by thesignificant attrition of patients by 12 months after surgerydue to transplantation, death, recovery, device exchange,and transfer of care.

Also, the sample size was small for patients withINTERMACS profiles 5 to 7 at 12 months after implant. Inaddition, data were missing. We prospectively assignedresponses for the VAS and 3 of 5 dimensions for patients whowere too sick to respond to the EQ-5D, which might havereduced overestimation of HRQOL across time. Sensitivityanalyses, without prospective assignment of missing data forpatients too sick to respond, supported our findings. We planto evaluate these analytic decisions in the future.

Conclusions

HRQOL is poor among patients with advanced heart failurewith INTERMACS profiles 1 to 7 before MCS implantationand improves to similar levels for all patients by 1 year aftersurgery. Monitoring HRQOL-related problems and devel-oping interventions may enhance HRQOL for all patients,regardless of disease severity.

Disclosure statement

K.L.G. is PI on an American Heart Association grant-in-aid. G.W.was supported by grants from the Alexander-von-HumboldtFoundation.

D.N. is a consultant for HeartWare and Thoratec Corporations.F.D.P. is National Co-PI for the ENDURANCE trial. None of theother authors has a financial relationship with a commercial entitythat has an interest in the subject of the presented manuscript orother conflicts of interest to disclose.

References

1. Braunwald E, Zipes D, Libby P, Bonow R, editors. Braunwald’s heartdisease: A textbook of cardiovascular medicine. Philadelphia: ElsevierInc; 2005.

2. Stevenson LW, Pagani FD, Young JB, et al. INTERMACS profiles ofadvanced heart failure: the current picture. J Heart Lung Transplant2009;28:535-41.

3. Kirklin JK, Naftel DC, Kormos R, et al. The fourth INTERMACSannual report: 4,000 implants and counting. J Heart Lung Transplant2012;31:117-36.

4. Stepanenko A, Potapov E, Jurmann B, et al. Outcomes of elective versusemergent permanent mechanical circulatory support in the elderly: asingle-center experience. J Heart Lung Transplant 2010;29:61-5.

Grady et al. QOL Improves After MCS 421

5. Boyle AJ, Ascheim DD, Russo MJ, et al. Clinical outcomes for continuous-flow left ventricular assist device patients stratified by pre-operativeINTERMACS classification. J Heart Lung Transplant 2011;30:402-7.

6. Kirklin JK, Naftel DC, Kormos RL, et al. Third INTERMACS annualreport: the evolution of destination therapy in the United States. J HeartLung Transplant 2011;30:115-23.

7. Grady KL, Meyer PM, Dressler D, et al. Longitudinal change in qualityof life and impact on survival after left ventricular assist deviceimplantation. Ann Thorac Surg 2004;77:1321-7.

8. Pagani FD, Miller LW, Russell SD, et al. Extended mechanicalcirculatory support with a continuous-flow rotary left ventricular assistdevice. J Am Coll Cardiol 2009;54:312-21.

9. Slaughter MS, Rogers JG, Milano CA, et al. Advanced heart failuretreated with continuous-flow left ventricular assist device. N Engl JMed 2009;361:2241-51.

10. Rogers J, Aaronson KD, Boyle AJ, et al. Continuous flow left ventri-cular assist device improves functional capacity and quality of life ofadvanced heart failure patients. J Am Coll Cardiol 2010;55:1826-34.

11. Field M, Cassel C, editors. Approaching death: improving care at the end oflife. Washington DC: Institute of Medicine, National Academy Press; 1997.

12. MacIver J, Rao V, Delgado D, et al. Choices: a study of preferences forend-of-life treatments in patients with advanced heart failure. J HeartLung Transplant 2008;27:1002-7.

13. Moskowitz A, Weinberg A, Oz M, Williams D. Quality of life with animplanted left ventricular assist device. Ann Thorac Surg 1997;64:1764-9.

14. Allen L, Stevenson L, Grady K, et al. Decision making in advancedheart failure: a statement from the American Heart Association.Circulation 2012;125:1928-52.

15. Baldwin T, Mann D. NHLBI’s program for VAD therapy for mod-erately advanced heart failure: The REVIVE-IT pilot trial. J CardiacFail 2010;16:855-8.

16. Spilker B. Quality of life and pharmacoeconomics in clinical trials. 2nded. New York: Lippincott Williams & Williams; 1996.

17. EuroQol group. EuroQol a new facility for the measurement of healthrelated quality of life. Health Policy 1990;16:199-208.

18. Rabin R, de Charro F. EQ-5D: a measure of health status from theEuroQol Group. Ann Med 2001;33:337-43.

19. van Agt H, Essink-Bot ML, Krabbe P, Bonsel G. Test-retest reliabilityof health state valuations collected with the EuroQoL questionnaire.Soc Sci Med 1994;39:1537-44.

20. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term mechanical leftventricular assistance for end-stage heart failure. N Engl J Med 2001;345:1435-43.

21. Park SJ, Tector A, Piccioni W, et al. Left ventricular assist devices asdestination therapy: a new look at survival. J Thorac Cardiovasc Surg2005;129:9-17.

22. Grady KL, Meyer P, Mattea A, et al. Improvement in quality of lifeoutcomes 2 weeks after left ventricular assist device implantation.J Heart Lung Transplant 2001;20:657-69.

23. Grady KL, Jalowiec A, White-Williams C. Quality of life 6 monthsafter heart transplantation compared with indicators of illness severitybefore transplantation. Am J Crit Care 1998;7:106-16.

24. Johnson J, Coons SJ. Comparison of the EQ5-D and SF-12 in an adultU.S. sample. Quality Life Res 1998;7:155-66.

25. Luo N, Johnson J, Shaw J, et al. Self-reported health status of thegeneral adult U.S. population as assessed by the EQ-5D and HealthUtilities Index. Medical Care 2005;43:1078-86.

26. Calvert MJ, Freemantle N, Cleland J. The impact of chronic heartfailure on health-related quality of life data acquired in the baselinephase of the CARE-HF study. Eur J Heart Fail 2005;7:243-51.

27. Almenar-Pertejo M, Almenar L, Martinez-Dolz L, et al. Studyon health-related quality of life in patients with advanced heart failurebefore and after transplantation. Transplant Proc 2006;38:2524-6.

28. Dyer M, Goldsmith K, Sharples L, Buxton M. A review of healthutilities using the EQ-5D in studies of cardiovascular disease. HealthQual Life Outcomes 2010;8:13.