Highlights
- •
Despite trending upward, the overall prevalence of palliative care usage is exceedingly low (14.1%).
- •
One third of all patients with advanced heart failure complicated by cardiogenic shock requiring mechanical circulatory support died in hospital.
- •
Amongst the palliative care cohort, we found a lower incidence of invasive procedures (such as pacemaker implantations, defibrillator implantations, insertion of percutaneous feeding tubes) and comparable complication rates (such as major bleeding and transfusion of any blood product) when compared with no palliative care use, in those who were discharged alive and those who died in hospital.
- •
Utilizing a multivariate logistic regression model: older age, female sex, coronary artery disease, and number of procedures performed at the indexed hospitalization were main predictors for palliative care consultation.
Prior studies have shown that the early inclusion of palliative care (PC) specialist is associated with better end-of-life experiences. The National Inpatient Sample Database was queried from 2012 to 2017 for relevant of ICD)-9 and -10 procedural and diagnostic codes to identify patients above 18 years with advanced heart failure (HF) admitted with cardiogenic shock (CS) requiring mechanical circulatory support (MCS). Baseline characteristics, utilization trends and invasive procedures and complications were compared among patients evaluated by PC and those who were not. There were 65,230 patients hospitalized for advanced HF complicated by CS requiring MCS, of these a PC consult was placed in in 9,200 patients (14.1%) and trended upward from 9.4 to 16.8%, between 2012 to 2017. The majority of patients, (37.3%) from the total population died in hospital. In reference to patients who were discharged alive, PC consultation was associated with a lower incidence of invasive procedures such as mechanical ventilation, pacemaker implantation, defibrillator implantation, insertion of percutaneous feeding tubes and tracheostomies performed (p <0.05 for all) whereas complications such as major bleeding, septic shock, transfusion of any blood product were comparable between both cohorts (nonsignificant p value for all). On the other hand, in those patients who died in hospital PC was associated with a lower incidence of pacemaker implantation, defibrillator implantation and insertion of percutaneous feeding tubes (p <0.05 for all). Despite the high morbidity and mortality associated with advanced HF patients with CS requiring MCS, the overall prevalence of PC consultation is exceedingly low. When utilized, the incidence of invasive procedures was lower. This study highlights the underutilization of PC services in this patient population, precluding any perceived benefit in end-of-life experiences.
With an aging United States population, the number of patients being diagnosed with heart failure (HF) and by extension advanced HF is increasing, with a median survival after the first HF-related hospitalization of only 2.4 years. Currently, there is a lack of robust evidence from large prospective trials for the use of mechanical circulatory support (MCS) strategies for advanced HF to prolong life, and these short-term MCS devices carry significant complications. , Although advanced HF patients presenting with cardiogenic shock (CS) may be stabilized with temporcgqz MCS devices, it is important to acknowledge the high burden of co-morbidities (cardiorenal syndrome/renal dysfunction, liver dysfunction/failure) that prevent candidacy for transplant or left ventricular assist device implantation in a patient population with a low-survival rate. Prior studies have shown that the early inclusion of palliative care (PC) is associated with better end-of-life experiences. Currently, North American and European HF societies recommend the early use of PC in advanced HF patients in order to provide care that is congruent with patient values, wishes, and preferences. We sought to evaluate the in-hospital utilization and outcomes of PC consultation in patients with advanced HF complicated by cardiogenic shock requiring MCS.
Methods
Data for analysis were collected from the National Inpatient Sample (NIS) provided by the Healthcare Cost and Utilization Project (HCUP) between 2012 and 2017. The NIS offers the largest database of hospitalizations, representing a 20% random, stratified, sample of hospital discharges in the United States and contains over 7 million hospital discharge data from about 1000 hospitals annually. To identify the study population, International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases, tenth revision, Clinical Modification (ICD-10-CM) diagnostic and procedural codes were used. Institutional review board approval was not needed, as all patient information is de-identified within the NIS. We identified all patients aged ≥ 18 admitted with advanced HF and cardiogenic shock who required MCS using ICD-9 and ICD-10 diagnostic codes. Our final study cohort comprised 13,046 patients with advanced HF and cardiogenic shock who required MCS, which, using NIS-provided trend discharge weights, corresponds to an estimated 65,230 overall nationwide hospitalizations during this time period. We separated patients into two cohorts, those in whom PC was consulted (see data supplement for ICD codes) and those in whom PC was not. We analyzed and reported outcomes in those who were discharged alive and those who died in hospital. All the data under the NIS are publicly available. Detailed methods used for statistical analyses are presented under the Data Supplement, which can be used for replication of our results. Baseline characteristics included in this study are listed in Table 1 . The primary outcomes of interest were the PC utilization and the number of invasive procedures during the hospitalization. Secondary outcomes of interest included the total cost of hospitalization, length of hospital stay, rates of mechanical ventilation, defibrillator implantation, cardiac surgery, major bleeding, tracheostomy procedures and discharge dispositions. The diagnostic codes of the complications are listed in the online supplemental appendix A. Data analysis was conducted following recommended methodological standards for the NIS . Univariate and multivariate logistic regression analyses were performed to identify independent predictors of in-hospital PC consultation. The Pearson Chi-Squared (χ 2 ) test was used for categorical variables, independent samples T-testing was used for parametric continuous variables and Mann-Whitney’s test for non-parametric continuous variables. A p-value of <0.05 was deemed statistically significant. All statistical analyses were performed using SPSS (IBM SPSS Statistics for Mac, Version 26.0. Armonk, NY: IBM Corp.).
| Covariates | Palliative Care Consultation | ||
|---|---|---|---|
| Yes (n=9200) | No (n=56030) | p-Value | |
| Age (years), mean (SD) | 66.12±14.6 | 63.6±14.8 | 0.853 |
| Women | 2855 (31.0%) | 16065 (28.7%) | <0.001 |
| White | 5970 (71.2%) | 36055 (68.8%) | <0.001 |
| Black | 1020 (12.2%) | 7030 (13.4%) | |
| Hispanic | 680 (8.1%) | 4515 (8.6%) | |
| Asian/Pacific Islander | 295 (3.5%) | 1840 (3.5%) | |
| Other | 365(4.4%) | 2615 (5.0%) | |
| PRIMARY EXPECTED PAYER | |||
| Medicare | 5550 (60.5%) | 30070 (53.7%) | <0.001 |
| Medicaid | 900 (9.8%) | 6270 (11.2%) | |
| Private Insurance | 2120 (23.1%) | 15550 (27.8%) | |
| Self-Pay | 350 (3.8%) | 2225 (4.0%) | |
| Other | 250 (2.7%) | 1680 (3.0%) | |
| MEDIAN HOUSEHOLD INCOME (Percentile) | |||
| 0-25 th | 2510 (27.8%) | 15275 (27.9%) | 0.463 |
| 26 th -50 th | 2290 (25.3%) | 14025 (25.6%) | |
| 51 st -75 th | 2305 (25.5%) | 13585 (24.8%) | |
| 76 th -100 th | 1930 (21.4%) | 11955 (21.8%) | |
| Mechanical Circulatory Support Device | |||
| Intra-aortic Balloon Pump | 6235 (67.8%) | 43655 (77.9%) | <0.001 |
| Percutaneous Ventricular Assist Devices (Impella and Tandem Heart) | 2170 (23.6%) | 9785 (17.5%) | <0.001 |
| Extracorporeal Membrane Support (ECMO) | 1890 (20.5%) | 6915 (12.3%) | <0.001 |
| Implantable Ventricular Assist Device | 1280 (13.9) | 5615 (10.0) | <0.001 |
| Combination of Percutaneous Ventricular Assist Devices (Impella and Tandem Heart), ECMO or Implantable Ventricular Assist Device | 4085 (44.4%) | 17365 (31.0%) | <0.001 |
| Elective Admission | 880 (9.6%) | 6580 (11.8%) | <0.001 |
| Diabetes Mellitus | 3325 (36.3%) | 21865 (39.2%) | <0.001 |
| Hypertension | 4525 (62.3%) | 26080 (62.1%) | 0.760 |
| Atrial Fibrillation | 3445 (37.4%) | 20065 (35.8%) | 0.002 |
| Coronary artery disease | 6280 (68.6%) | 39140 (70.2%) | 0.002 |
| Acute Coronary Syndrome | 5280 (58.0%) | 34220 (62.1%) | <0.001 |
| Prior Myocardial Infarction | 1105 (12.0%) | 5750 (10.3%) | <0.001 |
| Prior Percutaneous Coronary Intervention | 1000 (10.9%) | 5150 (9.2%) | <0.001 |
| Prior Coronary Artery Bypass Grafting | 530 (5.8%) | 3095 (5.5%) | 0.351 |
| Peripheral Arterial Disease | 945 (10.6%) | 5520 (10.1%) | 0.219 |
| Chronic Kidney Disease without Renal Replacement Therapy | 6430 (70.5%) | 37995 (68.3%) | <0.001 |
| End Stage Renal Disease on Hemodialysis | 295 (3.2%) | 2325 (4.1%) | <0.001 |
| Prior Stroke | 250 (2.7%) | 1125 (2.0%) | <0.001 |
| Chronic Obstructive Pulmonary Disease | 1415 (17.0%) | 9145 (17.6%) | 0.153 |
| Chronic Liver Disease | 720 (8.0%) | 3725 (6.8%) | <0.001 |
| Obesity | 1195 (13.0%) | 8790 (15.7%) | <0.001 |
| Smoker | 2475 (26.9%) | 16160 (28.8%) | <0.001 |
| Alcohol User | 430 (4.7%) | 2415 (4.4%) | 0.128 |
| Anemia | 505 (9.3%) | 2545 (8.7%) | 0.163 |
| Thrombocytopenia | 2635 (28.6%) | 14690 (26.2%) | <0.001 |
| Prior Malignancy | 475 (5.2%) | 2860 (5.1%) | 0.803 |
| Malnutrition | 170 (1.8%) | 580 (1.0%) | <0.001 |
| Do Not Resuscitate (DNR%) Order in Place | 2730 (29.7%) | 2555 (4.6%) | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
