If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address for reprints: Charlotta Lindvall, MD, PhD, Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, 450 Brookline Ave, LW670, Boston, MA 02215-5450.
Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, MassDivision of Palliative Medicine, Department of Medicine, Brigham and Women's Hospital, Boston, Mass
Department of Psychosocial Oncology and Palliative Care, Dana-Farber Cancer Institute, Boston, MassDivision of Palliative Medicine, Department of Medicine, Brigham and Women's Hospital, Boston, Mass
To assess baseline patient characteristics and identify factors associated with in-hospital mortality after ventricular assist device (VAD) placement.
Methods
Cross-sectional study using the National Inpatient Sample database from January 2010 to December 2014. Analyses were performed with sample weights provided by the National Inpatient Sample, which are reported ± the standard error of the mean.
Results
Weighted samples yielded 15,021 ± 1111 patients who received a VAD. The mean age at time of implantation was 56.6 years. Most recipients were white (59.9%) and male (75.0%). Among older patients, in-hospital mortality increased from 17.2% to 48.2% when 1 or more high-risk interventions (cardiac surgery, prolonged mechanical ventilation, hemodialysis, or extracorporeal membrane oxygenation) preceded VAD placement (P < .001). In comparison, in-hospital mortality in younger patients increased from 11.1% to 29.4% when 1 or more of these same procedures preceded VAD placement. The mortality difference associated with these procedures was 19% greater in older patients compared with younger patients (95% confidence interval [CI], 9%-28%). In-hospital mortality among VAD recipients was associated with age older than 65 years (odds ratio [OR], 1.76; 95% CI, 1.29-2.40), female sex (OR, 1.27; 95% CI, 0.97-1.67), and at least 1 high-risk intervention preceding VAD (OR, 5.52; 95% CI, 4.27-7.13).
Conclusions
Older patients who underwent 1 or more intensive treatments before VAD placement had a nearly 50% inpatient mortality and were unlikely to receive a cardiac transplantation. Refining patient selection might help better align VAD with those most likely to benefit.
We assessed outcomes after VAD placement in older adults. Older adults who underwent intensive treatments (cardiac surgery, mechanical ventilation, hemodialysis, or ECMO) before VAD implantation had a 48.2% mortality rate and are unlikely to receive cardiac transplantation. Refining patient selection might help better align treatment with those most likely to benefit.
Ventricular assist devices (VADs) have emerged as an alternative to transplantation for the >250,000 patients in the United States with end-stage heart failure.
Improved technology has made VAD, and especially left VAD (LVAD), implantation a treatment option for an increasing number of patients, particularly in older adults who are unlikely to receive a heart transplantation.
At the same time, the number of heart transplantations has remained stable and relatively uncommon in older adults. Currently, only 15% of heart transplantations are performed in adults older than the age of 65 and less than 2% in adults older than the age of 70 years.
Clinical strategies and outcomes in advanced heart failure patients older than 70 years of age receiving the HeartMate II left ventricular assist device.
Furthermore, increased frailty and comorbidities might attenuate the benefit of VAD therapy. VAD therapy requires substantial caregiving, which might be more challenging when the primary caregiver is another older adult. The decision to implant a VAD involves complex trade-offs, and can exert a great effect on patients and families.
An improved understanding of VAD implantation, including hospital outcomes might identify opportunities to optimize patient selection and thereby improve quality by lowering mortality, minimizing unnecessary suffering, increasing access, and reducing cost.
Previous studies have shown increased risk of early mortality after VAD implantation in patients older than the age of 65 years.
In addition, patients who require hemodialysis, intubation, or who undergo same-admission cardiac surgery before VAD placement have an increased risk of early mortality.
We hypothesized that in older patients, intensive treatments before VAD placement would result in a greater increase of in-hospital mortality compared with younger patients. We used the National Inpatient Sample (NIS) to compare younger and older patients receiving VADs over a 5-year period (2010-2014). Specifically, we analyzed: (1) demographic and clinical characteristics; (2) differences in hospital course, mortality, and disposition; and (3) temporal trends in the number of VADs placed. We captured and compared the effects of intensive treatments before VAD placement. These treatments included prolonged intubation (>96 hours), hemodialysis, intra-aortic balloon pump (IABP), same-admission cardiac surgery, and extracorporeal membrane oxygenation (ECMO).
Methods
Data Source
We used the NIS, which is the largest publicly available all-payer inpatient health care database in the United States, yielding national estimates of diagnoses, procedure utilization, and outcomes for hospital inpatient stays from a 20% stratified sample of US hospitals and is part of the Healthcare Quality and Utilization Project.
The database contains a nationally representative sample from more than 7 million hospitalizations annually. Applying sample weights provided by the NIS, the database projects estimates for more than 36 million hospitalizations annually. The NIS provides data on patient demographic characteristics, in-hospital clinical outcomes, hospital characteristics, and hospital charges. Federal hospitals are not included in the NIS. Quality control and validation of the NIS are performed by the Agency for Healthcare Research and Quality (AHRQ). For all demographic variables, except race/ethnicity, there is <1% missing data. Missing race/ethnicity was categorized as “unknown” in our results. Data from the NIS has been used to describe patterns of health care usage and outcomes of major procedures.
The database was provided with deidentified patient information and thus deemed exempt from institutional review by the Human Research Committee at Partners HealthCare.
Study Sample and Primary Outcomes
We included all individuals ≥18 years old who received a VAD during a hospitalization between January 1, 2010 and December 31, 2014. VAD was defined by the International Classification of Diseases, Ninth Revision, Procedural Coding System (ICD-9-PCS) code for VAD placement (37.66). This code is specific to implantation of a durable VAD and excludes temporary procedures such as pulsation balloons and percutaneous external heart assist devices as well as total internal biventricular heart replacement. This code has been stable in its usage throughout the study period. Use of this ICD-9-PCS code to identify patients who received a VAD has been previously validated in several studies.
We excluded patients who underwent a heart transplantation (ICD-9-PCS code 37.51) or combined heart-lung transplantation (ICD-9-PCS code 37.51) during same admission as VAD placement. The primary outcome of this study was in-hospital mortality. Secondary outcomes included length of hospital stay and hospital charges.
Covariates
Demographic covariates included age, sex, race/ethnicity, primary insurance type, and median income quartile on the basis of zip code. Race/ethnicity was reclassified as non-Hispanic white, non-Hispanic black, other (Hispanic, Asian, and Native American), and unknown. Other covariates included type of admission and geographic region (Northeast, Midwest, South, and West). Comorbidities were defined using International Classification of Diseases, Ninth Revision, Clinical Modification codes and the Elixhauser comorbidity index. The Elixhauser Index used in this study is a well validated indicator of comorbidities and was designed specifically to measure hospital charges, length of stay, and in-hospital mortality, which were the primary and secondary outcomes of this study. Furthermore, the Elixhauser Index was developed and validated using AHRQ data using the same variables that are reported in the NIS database (also maintained by AHRQ).
In this study, Elixhauser comorbidities were generated from International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes using AHRQ Comorbidity Software.
To determine the validity of dichotomization, we treated age as a continuous variable and used a logistic regression model to plot age against probability of in-hospital mortality after VAD implantation (Figure 1). We determined that mortality risk as a function of age is relatively stable until the age of 65 years, at which point it begins to dramatically increase. Because of the previous categorization of age within the literature and the findings in our regression we chose to categorize age into younger (18-65) and older (>65) categories.
Figure 1Plot of predicted probability of death after ventricular assist device placement in a logistic regression model with age as a continuous variable. Bars around central line represent upper and lower limits of 95% confidence intervals.
Circulatory support and co-therapy were identified using ICD-9 codes listed in Table E1. All hospital charges were adjusted for inflation with reference to 2014 US dollars, using the latest Consumer Price Index data (http://www.usinflationcalculator.com).
VAD surgery date and admission date were used to determine intensive procedures or treatments that occurred 1 hospital day before VAD placement. A 1 hospital day cutoff was used to ensure invasive treatments directly related to the VAD implantation (eg, mechanical ventilation during surgery) were not included in the analysis. Inpatient intensive procedures or treatments received during admission before VAD implantation were identified using ICD-9 procedure codes. These included IABP, cardiac surgery, mechanical ventilation, hemodialysis, and ECMO. In the case of cardiac surgery, we identified surgeries on the basis of ICD-9 procedure codes, which are listed in Table E1. We included open and endovascular procedures involving the valves and septa, coronary vessels, cardiotomy, pericardiotomy, and ascending aorta.
Statistical Analysis
All analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC). Frequencies, proportions, and 95% confidence intervals (CIs) were calculated and weighted to reflect national estimates using inverse sampling weights provided by the NIS. Trend weights were used for 2010-2011 NIS data to adjust for the redesign in 2012. χ2 Tests were used to compare the demographic and clinical characteristics. Continuous variables, such as length of stay and hospital charges, were compared using Wilcoxon rank sum test. Multivariable logistic regression was used to identify associations with in-hospital mortality. We used a model that included all patient demographic characteristics, hospital factors, and comorbidities measured using the Elixhauser comorbidity index. To avoid collinearity we combined preimplantation interventions with significant clinical overlap into a single high-risk intervention variable. Difference-in-difference analysis was used to compare the effects of preoperative interventions on mortality in older and younger patients. Finally, χ2 tests and Wilcoxon rank sum tests were used to compare temporal trends of VAD use, in-hospital mortality, and hospital charges from 2010 to 2014. A 2-tailed P < .05 was considered to indicate statistical significance.
Results
Study Population
Results are expressed as weighted samples ± the standard error of the mean. We identified 15,021 ± 1111 episodes of VAD implantation. The demographic characteristics, clinical characteristics, and hospital course of patients receiving a VAD are presented in Table 1. Most of the recipients were white (59.9%) and male (75.0%). The average score on the Elixhauser Comorbidity Index was 3.60 ± 0.07. Individual components of the Elixhauser Comorbidity Index are listed in Table E2. More than half of the patient population presented through the emergency department (65.1%) and a large percentage were covered either through Medicare (48.5%) or private insurance (36.1%). Overall, mortality was 12.9%; however, nearly half were able to be discharged home without services, or home with additional home care (49.8%). The average length of stay was 33.2 ± 0.68 days for patients who were alive at discharge and 39.0 ± 1.89 days for those who died in-hospital. Similarly, mean hospital charges were $767,431 ± $21,0.23 for patients alive at discharge and $1,037,622 ± $49,259 for those who died in-hospital. Finally, the most common preoperative procedure was IABP, which was performed in approximately 1 in 5 patients. Other major preoperative interventions including prolonged mechanical ventilation, ECMO, hemodialysis, and same-admission cardiac surgery occurred in <10% of the study population.
Includes 5 major groups of surgeries operations on valves and septa, vessels of heart, cardiotomy and pericardiotomy, pericardiotomy and excision of lesion of heart, and major aortic dissection or aneurysm repair.
557 (3.7)
86
392 (3.6)
52
165 (3.9)
34
.62
Mechanical ventilation
1250 (8.3)
151
969 (9.0)
104
281 (6.6)
47
<.001
ECMO
526 (3.5)
82
431 (4.0)
61
95 (2.2)
21
<.001
Hemodialysis
575 (3.8)
96
332 (3.2)
52
243 (5.1)
44
.01
SE, Standard error of the mean of the weighted sample; VAD, ventricular assist device; IABP, intra-aortic balloon pump; ECMO, extracorporeal membrane oxygenation.
∗ Includes Hispanic, Asian or Pacific Islander, Native American, and other.
† Includes self-pay, no charge, and other.
‡ Includes 5 major groups of surgeries operations on valves and septa, vessels of heart, cardiotomy and pericardiotomy, pericardiotomy and excision of lesion of heart, and major aortic dissection or aneurysm repair.
In stratified analysis, all differences between the age groups were statistically significant (Table 1). Compared with patients ages 18-65 years, older VAD recipients were more likely to be white (72.4% vs 55.0%; P < .001) and male (84.2% vs 73.4%; P < .001). Older VAD recipients had a higher Elixhauser index score (3.80 vs 3.53; P < .001). Regional utilization differences also existed, with a higher proportion of older patients receiving a VAD in the West (16.8% vs 13.2%; P < .001) and Northeast (22.7% vs 19.0%; P < .001). Patients ages 18-65 years were more likely to have been admitted through the emergency department (66.4% vs 62.1%; P = .03). The mean length of stay was >30 days in both groups. Patients ages 18-65 years who died in-hospital had a longer mean length of stay than patients older than 65 years who died in-hospital (41.9 vs 34.3 days; P = .03). Total hospital charges were similar in both groups. In both age groups, a significant proportion of patients underwent invasive treatments before VAD placement, which included IABP, cardiac surgery, mechanical ventilation >96 hours, ECMO, and hemodialysis. Among these treatments patients between the ages of 18 and 65 years tended to have a high rate of prolonged mechanical ventilation (9.0% vs 6.6%; P < .001) and use of ECMO (4.0% vs 2.2%; P < .001). Conversely, those aged 18-65 years had a significantly lower rate of hemodialysis (3.2% vs 5.1%; P = .01).
Preoperative Interventions and Mortality
In-hospital mortality for patients older than 65 years was significantly higher than in patients ages 18-65 years (17.2% vs 11.1%; P < .001). In-hospital mortality among patients who received intensive procedures/treatments during the index admission before VAD implantation surgery, stratified according to age, is shown in Table 2 and Figure 2. The 4 procedures/treatments most highly correlated with inpatient mortality were same-admission cardiac surgery, hemodialysis, mechanical ventilation, and ECMO. Among patients older than 65 years, 649 ± 83 (15.3%) had at least 1 of these procedures, before VAD implantation. Of these patients, 313 ± 52 (48.2%) died during hospitalization. For patients ages 18-65 years, 1694 ± 169 (15.7%) had 1 of these 4 procedures before VAD implantation. Of these patients, 498 ± 65 (29.4%) died during hospitalization. Although having 1 of these procedures increased mortality in younger and older patients, there was a significantly greater increase in mortality in patients older than 65 years.
Table 2Difference-in-difference analysis of mortality according to intervention in patients before receipt of VAD
Includes 5 major groups of surgeries: operations on valves and septa, vessels of heart, cardiotomy and pericardiotomy, pericardiotomy and excision of lesion of heart, and major aortic dissection or aneurysm repair.
High-risk intervention defined by cardiac surgery, mechanical ventilation (>96 hours), ECMO, or hemodialysis before VAD placement, all of which are associated with high in-hospital mortality.
498 (29.4)
65
313 (48.2)
52
19% (9.0-28)
Baseline in-hospital mortality after VAD implantation
1204 (11.1)
120
729 (17.2)
81
6.1% (+5.3-7.1)
Mortality difference is the change in mortality between each age group in those who received the indicated treatment. SE, Standard error of the mean; CI, confidence interval; IABP, intra-aortic balloon pump; ECMO, extracorporeal membrane oxygenation; VAD, ventricular assist device.
∗ Includes 5 major groups of surgeries: operations on valves and septa, vessels of heart, cardiotomy and pericardiotomy, pericardiotomy and excision of lesion of heart, and major aortic dissection or aneurysm repair.
† High-risk intervention defined by cardiac surgery, mechanical ventilation (>96 hours), ECMO, or hemodialysis before VAD placement, all of which are associated with high in-hospital mortality.
In our multivariable logistic regression model the 4 highest risk interventions (same-admission cardiac surgery, hemodialysis, mechanical ventilation >96 hours, and ECMO) were combined into a high-risk treatment category to avoid collinearity. These high-risk treatments were associated with increased in-hospital mortality (odds ratio [OR], 5.52; 95% CI, 4.27-7.13). Univariate analysis is reported separately in Table E3. Age older than 65 years was also associated with increased in-hospital mortality (OR, 1.77; 95% CI, 1.29-2.40; Table 3). However, we did not see a significant interaction effect between age older than 65 years and high-risk treatments in multivariate analysis (P = .24). Elective admission (as opposed to admission through the emergency department) was associated with lower odds of mortality (OR, 0.75; 95% CI, 0.57-0.98). In multivariate regression, IABP was not associated with a significant change in overall mortality (OR, 0.91; 95% CI, 0.69-1.20). Because of the categorization of age we performed a sensitivity analysis using age as a continuous variable. There were no changes to the major trends and minimal change in the reported ORs.
Table 3Multivariate analysis of in-hospital mortality after the VAD procedure and sensitivity analysis using age as a continuous variable
Includes vasopressors, intra-aortic balloon pump, extracorporeal membrane oxygenation, hemodialysis, mechanical ventilation, and same admission cardiac surgery before VAD placement.
5.52 (4.27-7.13)
5.65 (4.35-7.34)
Because of the nonlinear relationship of age and mortality age and age2 were included in the continuous model. The results of mortality as a plot of age against the probability of in-hospital mortality is shown in Figure 1. Data are presented as OR (95% CI). IABP, Intra-aortic balloon pump; VAD, ventricular assist device.
∗ Includes Hispanic, Asian or Pacific Islander, Native American, and other.
† Includes self-pay, no charge, and other.
‡ Includes vasopressors, intra-aortic balloon pump, extracorporeal membrane oxygenation, hemodialysis, mechanical ventilation, and same admission cardiac surgery before VAD placement.
Temporal Trends in Heart Transplantation and VAD Placement According to Age
The number of VAD implantations per year in patients ages 18-65 years increased from 1808 ± 459 in 2010 to 2560 ± 317 in 2014 (Figure 3). The number of VAD implantations per year in patients older than 65 years more than doubled with 532 ± 143 in 2010 to 1100 ± 120 in 2014. In-hospital mortality during this period did not significantly change on the basis of age. The rate of heart transplantations declined steadily with age. Of nearly 10,000 heart transplantations, only 1515 (15.1%) were performed in individuals older than the age of 65 years (Figure 4). By the age of 71 years, this number decreased to 229 (2.2%), indicating that most of the 2025 ± 198 patients with a VAD older than the age of 71 years were unlikely to receive heart transplantation.
Figure 3Ventricular assist device implantation and in-hospital mortality between 2010 and 2014 for those aged 18-65 years and those older than 65 years.
Figure 4Distribution of ventricular assist device (VAD) placement and heart transplantation stratified according to age. After the age of 70 years the opportunity for heart transplantation is exceptionally rare.
The objective of this study was to analyze the baseline demographic characteristics and characteristics of patients receiving VADs and determine the effect of preimplantation interventions on in-hospital mortality. We found that in-hospital mortality after VAD placement begins to rapidly increase in patients older than the age of 65 years. Although previous studies have reported increasing age to be a risk factor for early mortality,
our study shows an explanation for this finding. Intensive procedures before VAD placement resulted in a nearly 50% in-patient mortality in older patients. Finally, this work confirmed a low rate of cardiac transplantations in patients older than the age of 65 years, indicating that most VAD placements in this older age group was destination therapy rather than a bridge to transplantation (Graphical Abstract). The high rate of in-patient mortality in older adults who have intensive procedures before VAD placement, coupled with the unlikelihood of cardiac transplantation, is an important distinction and deserves recognition by clinicians and discussion with patients (Video 1).
we chose to further validate this decision. Using linear regression, we showed an exponential increase of in-hospital mortality after VAD placement in individuals older than the age of 65 years. Clinically, the age of 65 years represents the time point at which the rate of cardiac transplantation begins to decline. Although in younger patients high-risk VAD placement might be attempted to bridge the patient to cardiac transplantation, in older patients transplantation is unlikely and VADs must often be considered destination therapy.
Previous studies have shown similar short-term mortality between younger and older VAD recipients. The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) report showed 30-day survival of 93% for VAD patients aged 70 years or older.
Clinical strategies and outcomes in advanced heart failure patients older than 70 years of age receiving the HeartMate II left ventricular assist device.
Clinical strategies and outcomes in advanced heart failure patients older than 70 years of age receiving the HeartMate II left ventricular assist device.
In our study 17.2% of patients aged older than 65 years died in the hospital after VAD implantation. The discrepancy in short-term mortality might in part be explained by differences in measurement. A subset of patients captured in the NIS died >30 days after VAD implantation, but while still in-hospital. The high mortality identified in our study could also stem from differences in cohort selection and study periods. INTERMACS collects data from sites approved by the registry, which currently includes 127 centers. In our study, a cohort was extracted on the basis of a weighted sample of hospitals participating in the NIS, which in 2011 included approximately 1000 centers and is the single largest all-payer publicly available database in the United States. The inclusion of hospitals outside of INTERMACS helps explain the higher overall number of VAD placements seen in this study compared with recent reports using the INTERMACS registry. The INTERMACS registry reports a 1-year mortality rate of 20% in patients after LVAD implantation. Other multi-institutional studies have shown a 1-year mortality of 25% in patients aged 70 years and older compared with 16% in patients aged 60-69 years.
Both are higher than the in-hospital mortality reported in this study, but compare well if our data are projected out.
The number of patients older than age 65 years with heart failure is expected to increase to 3.9 million by the year 2030. However, the availability of donor hearts remains stable; thus, we expect the utilization of VADs to increase.
Although VADs offer a more readily available method of cardiovascular support than heart transplantation, they are associated with significant morbidity, caregiver burden, and complicated end of life care.
These challenges place further importance on optimizing patient selection and the shared decision-making process before VAD implantation.
The decision to proceed with VAD placement remains a difficult challenge for clinicians. In particular, compared with heart transplantation, the criteria for VAD patient selection are less well defined.
Our study further suggests that the intensity of the immediate medical care should be included in the risk assessment of VAD surgery. Not surprisingly, intensive procedures such as cardiac surgery, ECMO, mechanical ventilation, or hemodialysis before VAD implantation were associated with high in-hospital mortality. This is true for older and younger patients, yet the increase is substantially higher in patients older than the age of 65 years. In our study, we found that nearly 50% of older patients who had 1 or more intensive procedures before VAD placement died in the hospital. In contrast, IABP had only a minor effect on in-hospital mortality in younger patients and in patients older than 65 years of age it had no significant effect on in-hospital mortality.
Finally, equitable access to this treatment remains a concern.
Older VAD recipients were overwhelmingly white and male. Historical evidence shows decreased access to evidence-based treatment for women and minority patients.
Further analysis on access will enable us to address potential disparities and cost considerations as VAD becomes more common.
Limitation
The findings of this study should be interpreted within the limitations of study design. First, clinical data such as INTERMACS profile, laboratory data, medications, and echocardiography data were not available, limiting our ability to study granular clinical characteristics. The data represent a cross-sectional assessment of device implantation; longitudinal clinical outcomes might offer a more comprehensive understanding of differences in clinical outcomes between patients ages 18-65 years versus patients older than 65 years. For example, we are unable determine readmission rates or long-term complications such as pump thrombosis. In our analysis, we used sample weights that are projected from NIS collected data. This method is designed to help study relatively rare events, such as VAD placement, and to identify national trends. This method has been validated several times, including in the study of VAD and congestive heart disease.
Although most VADs are left-sided, it is not possible to identify the number of right VADs implanted in this study. On the basis of INTERMACS data from this time period there were only 116 right-sided VADs placed between 2006 and 2014.
Although it is unlikely that all of these right VADs were implanted during our study period of 2010-2014, if they were they would only represent 0.8% of the study population. It is also not possible to identify the type or model of VAD device implanted. Furthermore, NIS data do not distinguish between VADs that were used for destination therapy versus bridge to transplantation because there are no administrative codes for that distinction.
We analyzed the frequency of heart transplantations in the NIS database and determined that only 229 patients older than 70 years received a heart transplantation between 2010 and 2014; thus, most VADs for patients older than 70 years were for destination therapy. We chose to investigate outcomes of patients between the years of 2010 and 2014, because this represents a period of relatively stability in administrative coding and allowed for more robust analysis. However, the field advances rapidly. Although data in this study are recent, it is possible they do not perfectly reflect contemporary trends. Finally, we are limited in ascertainment of true cost of hospital care accrued during the index admission; our study only used hospital charges as a proxy measure.
Conclusions
In this nationally representative cohort, older patients who underwent major procedures during the hospitalization before VAD implantation had high in-hospital mortality (48.2%) and a low likelihood of cardiac transplantation. Future efforts to improve patient selection might help better align destination VAD implantation with those most likely to benefit.
Conflict of Interest Statement
Authors have nothing to disclose with regard to commercial support.
The authors thank Dr Steven Worthington for help with statistical analyses. The authors also thank the Institute for Quantitative Social Science at Harvard University for providing access to cluster computing.
Appendix
Table E1In-hospital intensive procedures and treatments
Description
ICD-9 code
Intra-aortic balloon pump
37.61
Mechanical ventilation
96.70, 96.71, 96.72
Hemodialysis
39.95
ECMO
37.62
Invasive mechanical ventilation ≥96 hours
96.72
Cardiac surgery
Operations on valves and septa of heart
Closed heart valvotomy, unspecified valve
35
Closed heart valvotomy, aortic valve
35.01
Closed heart valvotomy, mitral valve
35.02
Closed heart valvotomy, pulmonary valve
35.03
Closed heart valvotomy, tricuspid valve
35.04
Endovascular replacement of aortic valve
35.05
Transapical replacement of aortic valve
35.06
Endovascular replacement of pulmonary valve
35.07
Transapical replacement of pulmonary valve
35.08
Endovascular replacement of unspecified heart valve
35.09
Open heart valvuloplasty without replacement, unspecified valve
35.1
Open heart valvuloplasty of aortic valve without replacement
35.11
Open heart valvuloplasty of mitral valve without replacement
35.12
Open heart valvuloplasty of pulmonary valve without replacement
35.13
Open heart valvuloplasty of tricuspid valve without replacement
35.14
Open and other replacement of unspecified heart valve
35.2
Open and other replacement of aortic valve with tissue graft
35.21
Open and other replacement of aortic valve
35.22
Open and other replacement of mitral valve with tissue graft
35.23
Open and other replacement of mitral valve
35.24
Open and other replacement of pulmonary valve with tissue graft
35.25
Open and other replacement of pulmonary valve
35.26
Open and other replacement of tricuspid valve with tissue graft
35.27
Open and other replacement of tricuspid valve
35.28
Operations on papillary muscle
35.31
Operations on chordae tendineae
35.32
Annuloplasty
35.33
Infundibulectomy
35.34
Operations on trabeculae carneae cordis
35.35
Operations on other structures adjacent to valves of heart
35.39
Enlargement of existing atrial septal defect
35.41
Creation of septal defect in heart
35.42
Repair of unspecified septal defect of heart with prosthesis
35.5
Repair of atrial septal defect with prosthesis, open technique
35.51
Repair of atrial septal defect with prosthesis, closed technique
35.52
Repair of ventricular septal defect with prosthesis, open technique
35.53
Repair of endocardial cushion defect with prosthesis
35.54
Repair of ventricular septal defect with prosthesis, closed technique
35.55
Repair of unspecified septal defect of heart with tissue graft
35.6
Repair of atrial septal defect with tissue graft
35.61
Repair of ventricular septal defect with tissue graft
35.62
Repair of endocardial cushion defect with tissue graft
35.63
Other and unspecified repair of unspecified septal defect of heart
35.7
Other and unspecified repair of atrial septal defect
35.71
Other and unspecified repair of ventricular septal defect
35.72
Other and unspecified repair of endocardial cushion defect
35.73
Total repair of tetralogy of Fallot
35.81
Total repair of total anomalous pulmonary venous connection
35.82
Total repair of truncus arteriosus
35.83
Total correction of transposition of great vessels, not elsewhere classified
35.84
Interatrial transposition of venous return
35.91
Creation of conduit between right ventricle and pulmonary artery
35.92
Creation of conduit between left ventricle and aorta
35.93
Creation of conduit between atrium and pulmonary artery
35.94
Revision of corrective procedure on heart
35.95
Percutaneous balloon valvuloplasty
35.96
Percutaneous mitral valve repair with implant
35.97
Other operations on septa of heart
35.98
Other operations on valves of heart
35.99
Operations on vessels of heart
Open chest coronary artery angioplasty
36.03
Intracoronary artery thrombolytic infusion
36.04
Insertion of non–drug-eluting coronary artery stent(s)
36.06
Insertion of drug-eluting coronary artery stent(s)
36.07
Other removal of coronary artery obstruction
36.09
Aortocoronary bypass for heart revascularization, not otherwise specified
36.1
(Aorto)coronary bypass of 1 coronary artery
36.11
(Aorto)coronary bypass of 2 coronary arteries
36.12
(Aorto)coronary bypass of 3 coronary arteries
36.13
(Aorto)coronary bypass of 4 or more coronary arteries
36.14
Single internal mammary-coronary artery bypass
36.15
Double internal mammary-coronary artery bypass
36.16
Abdominal-coronary artery bypass
36.17
Other bypass anastomosis for heart revascularization
36.19
Heart revascularization by arterial implant
36.2
Open chest transmyocardial revascularization
36.31
Other transmyocardial revascularization
36.32
Endoscopic transmyocardial revascularization
36.33
Percutaneous transmyocardial revascularization
36.34
Other heart revascularization
36.39
Repair of aneurysm of coronary vessel
36.91
Other operations on vessels of heart
36.99
Cardiotomy and pericardiotomy
Incision of heart, not otherwise specified
37.1
Cardiotomy
37.11
Pericardiotomy
37.12
Pericardiectomy and excision of lesion of heart
Pericardiectomy
37.31
Excision of aneurysm of heart
37.32
Excision or destruction of other lesion or tissue of heart, open approach
37.33
Excision or destruction of other lesion or tissue of heart, endovascular approach
37.34
Partial ventriculectomy
37.35
Excision, destruction, or exclusion of left atrial appendage
37.36
Excision or destruction of other lesion or tissue of heart, thoracoscopic approach
37.37
Major aortic dissection or aneurysm repair
Unspecified site
441
Ascending aorta
441.01
Thoracoabdominal
441.03
Ascending aorta aneurysm, ruptured
441.1
Ascending aorta aneurysm without mention of rupture
441.2
Thoracoabdominal aneurysm, ruptured
441.6
Thoracoabdominal aneurysm, without mention of rupture
441.7
Aortic aneurysm of unspecified site without mention of rupture
441.9
ICD-9, International Classification of Diseases, Ninth Revision; ECMO, extracorporeal membrane oxygenation.
Note that the diagnosis of congestive heart failure and valvular heart disease are primary diagnoses for a large percentage of patients in this study and thus not readily captured by our Elixhauser comorbidity index. Because of the ubiquity of these diagnoses in the study population they are essentially dropped in this analysis.
Note that the diagnosis of congestive heart failure and valvular heart disease are primary diagnoses for a large percentage of patients in this study and thus not readily captured by our Elixhauser comorbidity index. Because of the ubiquity of these diagnoses in the study population they are essentially dropped in this analysis.
30 (0.2)
16
25 (0.2)
15
5 (0.1)
5
.54
Pulmonary circulation disorders
44 (0.3)
21
39 (0.4)
21
5 (0.1)
5
.29
Peripheral vascular disorders
1325 (8.8)
121
793 (7.4)
90
532 (12.6)
61
<.001
Hypertension
6628 (44.1)
518
4568 (42.3)
375
2060 (48.7)
179
.03
Paralysis
359 (2.4)
47
264 (2.5)
40
94 (2.2)
21
.71
Other neurologic disorders
783 (5.2)
90
521 (4.8)
69
261 (6.2)
39
.12
Chronic pulmonary disease
2715 (18.1)
241
1817 (16.8)
178
898 (21.2)
98
.02
Diabetes, uncomplicated
3928 (26.1)
301
2739 (25.4)
225
1188 (28.1)
111
.15
Diabetes, complicated
1085 (7.2)
102
722 (6.7)
79
363 (8.6)
47
.08
Hypothyroidism
1744 (11.6)
145
1108 (10.2)
106
636 (15.0)
68
<.001
Chronic kidney disease
5989 (39.9)
481
3905 (36.1)
330
2084 (49.2)
189
<.001
Liver disease
533 (3.5)
62
406 (3.8)
52
127 (3.0)
27
.31
Peptic ulcer disease, excluding bleeding
5 (0.03)
5
5 (0.04)
5
0 (0)
0
1
Acquired immune deficiency syndrome
20 (0.13)
10
20 (0.19)
10
0 (0)
0
1
Lymphoma
167 (1.1)
34
109 (1.0)
26
58 (1.4)
20
.4
Metastatic cancer
35 (0.2)
13
20 (0.18)
10
15 (0.4)
9
.37
Solid tumor without metastasis
140 (0.9)
31
85 (0.8)
22
55 (1.3)
19
.22
Rheumatoid arthritis/collagen vascular diseases
178 (1.2)
32
114 (1.1)
25
64 (1.5)
18
.29
Coagulopathy
5778 (38.5)
482
4010 (37.2)
360
1768 (41.8)
152
.02
Obesity
2431 (16.2)
187
2058 (19.1)
167
373 (8.8)
51
<.001
Weight loss
4102 (27.3)
346
2721 (25.2)
250
1381 (32.6)
130
.002
Fluid and electrolyte disorders
9778 (65.1)
759
7192 (66.7)
577
2586 (61.1)
215
.007
Blood loss anemia
255 (1.7)
44
172 (1.6)
35
83 (2.0)
22
.49
Deficiency anemias
3169 (21.3)
315
2310 (21.4)
242
886 (20.9)
96
.76
Alcohol use
385 (2.6)
50
336 (3.1)
45
49 (1.1)
17
.003
Drug abuse
330 (2.2)
49
321 (3.0)
48
9 (0.2)
7
<.001
Psychoses
483 (3.2)
66
389 (3.6)
61
94 (2.2)
23
.08
Depression
1639 (10.9)
153
1233 (11.4)
119
406 (9.6)
57
.14
SE, Standard error of the mean.
∗ Note that the diagnosis of congestive heart failure and valvular heart disease are primary diagnoses for a large percentage of patients in this study and thus not readily captured by our Elixhauser comorbidity index. Because of the ubiquity of these diagnoses in the study population they are essentially dropped in this analysis.
Includes vasopressors, IABP, extracorporeal membrane oxygenation, hemodialysis, mechanical ventilation, and same admission cardiac surgery before VAD placement.
5.50 (4.38-6.91)
<.001
Age × Age was used in addition to Age as a continuous variable because of the nonlinear relationship of age and mortality. OR, Odds ratio; CI, confidence interval; IABP, intra-aortic balloon pump; VAD, ventricular assist device.
∗ Includes Hispanic, Asian or Pacific Islander, Native American, and other.
† Includes self-pay, no charge, and other.
‡ Includes vasopressors, IABP, extracorporeal membrane oxygenation, hemodialysis, mechanical ventilation, and same admission cardiac surgery before VAD placement.
Clinical strategies and outcomes in advanced heart failure patients older than 70 years of age receiving the HeartMate II left ventricular assist device.
Dr. Lindvall was supported by a Junior Faculty Career Development Award from the National Palliative Care Research Center, New York, New York and a 2016 Pilot Award (NINR U24) from the Palliative Care Research Cooperative Group, Denver, Colorado.
The largest growth in durable left ventricular assist device (LVAD) therapy has arisen in patients ineligible for heart transplantation who receive the LVAD for destination therapy or permanent use. Patients receiving destination therapy are older and have more comorbidities, such as diabetes, renal dysfunction, and frailty, and are less resilient to recover from major complications of the therapy or achieve satisfactory improvement in quality of life.1,2 This older population presents immense challenges to selecting appropriate candidates for this therapy.
Congestive heart failure is a common pathology that increases with age. Medical treatment of patients with congestive heart failure offers limited survival and poor chance of functional recovery. Ventricular assist devices (VADs), mostly the last generation, may offer significant benefits in patients with refractory heart failure. Nevertheless, although heart transplantation (HT) has an acceptable long-term survival and represents the gold standard treatment for such patients, the number of HT procedures has not increased during the past decade.
32962-3&id=fx1.jpg){kind=link}
32962-3&id=fx2.jpg){kind=link}
32962-3&id=gr1.jpg){kind=link}
32962-3&id=gr2.jpg){kind=link}
32962-3&id=gr3.jpg){kind=link}
32962-3&id=gr4.jpg){kind=link}
32962-3&id=fx3.jpg){kind=link}