Palliative Care Consultation in Cardiogenic Shock Requiring Short-Term Mechanical Circulatory Support: A Retrospective Cohort Study

Abstract

Background:

Little is known about palliative care consultation (PCC) for patients with cardiogenic shock requiring short-term mechanical circulatory support (STMCS).

Objective:

To describe the utilization of PCC in this population.

Design:

Retrospective cohort study in a university medical center intensive care unit (ICU).

Setting/Participants:

In total, 195 patients aged >18 years with cardiogenic shock requiring STMCS were included. The cohort was divided into three categories: no PCC, early PCC (within seven days of STMCS), and late PCC (eight or more days after STMCS). Follow-up occurred during the index hospitalization.

Results:

Mean age was 59.3 ± 13.9 years; 67.9% were men. Mean follow-up period was 33.8 ± 37.7 days. Overall inpatient mortality was 52.3%. Ninety-four patients (48.2%) received PCC; 49 (25.1%) and 45 (23.1%) received early and late PCCs, respectively. STMCS duration, ICU stay after STMCS, and hospital stay after STMCS were significantly shorter in the no PCC group than the early PCC group (4 vs. 12 days, p < 0.001; 11 vs. 19 days, p = 0.004; and 16 vs. 19 days, p = 0.031; respectively). ICU stay after STMCS and hospital stay after STMCS were significantly shorter in the early PCC group than the late PCC group (19 vs. 38 days, p < 0.001; 19 vs. 49 days, p < 0.001; respectively). However, time from initial PCC to discharge was not significantly different between early and late PCC groups (18 vs. 31 days, p = 0.13).

Conclusions:

PCC was utilized in almost half of patients with cardiogenic shock requiring STMCS. PCC tends to occur toward the end of life regardless of the duration of STMCS. The optimal PCC timing remained unclear.

Introduction

Despite advances in the treatment of cardiogenic shock, short-term mortality remains unacceptable, as high as 40–60%.^1–3 Short-term mechanical circulatory support (STMCS), such as extracorporeal membrane oxygenation (ECMO) or use of a short-term ventricular assist device (STVAD), is able to provide a great deal of circulatory support to extremely sick patients and its use has been increasing rapidly,⁴ but mortality remains high.

Palliative care is an interdisciplinary approach that improves quality of life for patients with serious illness, as well as for their families, through prevention and relief of suffering. Palliative care includes assessment and treatment of physical symptoms, identification and relief of spiritual distress, expert communication to establish goals of care, assistance with complex decision making, and coordination of care among different health care specialities.⁵

Offering palliative care services in the intensive care unit (ICU) has been associated with better outcomes both for patients and families, including better symptom management, reduced ICU and hospital length of stay,^6–9 less use of unwanted life-prolonging treatments,^10,11 and significant cost savings.^12,13 It also has positive impacts on families in terms of increased family satisfaction¹⁴; reduced conflict with providers¹⁵; and alleviation of symptoms of anxiety, depression, and post-traumatic stress disorder.¹⁶

However, little is known about palliative care consultation (PCC) in this population, especially the timing of consultation. We hypothesize that PCC is not utilized in a timely manner in patients with cardiogenic shock who require STMCS. The purpose of this study is to describe the utilization of PCC in this patient population and to examine the relationship between the timing of PCC and patient outcomes.

Methods

This retrospective cohort study was approved by the Institutional Review Board of the Columbia University Medical Center (AAAE1866), and participants or their surrogates provided written informed consent for inclusion in a cardiogenic shock database and participation in the study. All patients in the ICU at Columbia University Medical Center aged 18 years or older who developed cardiogenic shock requiring STMCS (either ECMO or a STVAD) during 2014–2016 were eligible for inclusion. Patients with primary graft dysfunction after heart transplant and patients who previously received a durable left ventricular assist device (LVAD) were excluded.

Palliative care consultation

The palliative care team consists of board-certified palliative care physicians and nurse practitioners, social workers, and chaplains. Once consulted, the palliative care team evaluates and comanages symptoms with the ICU team; facilitates interdisciplinary goals-of-care conversations as needed; and provides psychosocial and spiritual support to patients, families of patients, and caregivers.

To examine the timing of PCC as well as the overall utilization of PCC, we divided the cohort into three categories: no PCC, early PCC (PCC within seven days of implementing STMCS), and late PCC (PCC eight or more days after implementing STMCS). Patients were followed up until death or discharge during the index hospitalization.

Outcome measures

The primary outcomes were overall utilization of palliative care in this population and the number of palliative care team visits. Palliative care team visits were defined based on palliative care team notes in patients' electronic medical records. If there were two or more notes by the palliative care team on the same day, it was counted as one visit. When STMCS was discontinued because of cardiac recovery, palliative care team visits after STMCS were also reviewed.

The secondary outcomes were days on STMCS, days in the ICU after implementing STMCS, days in the hospital after implementing STMCS, days after initial PCC, indications for STMCS, withdrawal of STMCS, and patient mortality. These outcomes were also reviewed separately for patients who died during the index hospitalization.

Statistical analysis

Chi-squared tests, the Mann-Whitney U test, and analysis of variance (ANOVA) were used to compare outcome measures between groups, as appropriate. All tests were two-sided, and p-values of <0.05 were considered to indicate statistical significance. Statistical analyses were performed using SPSS Statistics for Mac, version 24.0 (IBM, Armonk, NY).

Results

Out of 248 eligible patients, 53 were excluded because of primary graft dysfunction after heart transplant or previous history of LVAD therapy. Written consent was obtained from all remaining patients or their surrogates, so 195 patients were included in this analysis. Of those, 102 patients (52.3%) died before discharge. Mean age (SD) of all patients was 59.3 (±13.9) years; 132 patients (67.7%) were men. Mean follow-up period (SD) was 33.8 (±37.7) days. Ninety-four patients (48.2%) utilized PCC; of these, 49 patients (25.1%) fell into the early PCC group and 45 patients (23.1%) fell into the late PCC group. There was no difference in terms of body mass index or past medical history among the groups.

In terms of indications for STMCS, postcardiotomy shock (PCS) and acute myocardial infarction (AMI) were more common in the no PCC group, as opposed to AMI in the early PCC group and acute decompensated heart failure (ADHF) in the late PCC group (Table 1); the difference between the groups was significant (p = 0.048). The median number of palliative care visits during STMCS was similar between the early PCC and late PCC groups (p = 0.97); more PCC visits occurred after STMCS in the late PCC group, but the difference was not significant (p = 0.24).

Table 1.

Characteristics of Patients, Indications for and Withdrawal of Mechanical Circulatory Support, Palliative Care Consultations, and Mortality

	No PCC	Early PCC	Late PCC	p
	n = 101	n = 49	n = 45	p
Age, years (mean ± SD)	60.9 ± 13.9	59.7 ± 12.4	55.3 ± 15.2	0.08
Men, n (%)	64 (63.4)	37 (75.5)	31 (68.9)	0.323
BMI, kg/m², (mean ± SD)	28.9 ± 5.9	29.0 ± 5.9	29.0 ± 6.0	0.987
PMH
Hypertension, n (%)	71 (70.3)	32 (65.3)	28 (62.2)	0.599
Diabetes mellitus, n (%)	30 (29.7)	15 (30.6)	14 (31.1)	0.984
Cardiovascular accident, n (%)	6 (5.9)	4 (8.2)	5 (11.1)	0.551
Coronary artery disease, n (%)	53 (52.5)	31 (63.3)	27 (60.0)	0.408
Hyperlipidemia, n (%)	51 (50.5)	22 (44.9)	22 (48.9)	0.813
Indication for MCS, n (%)				0.048
PCS	33 (32.7)	9 (18.4)	13 (28.9)
AMI	31 (30.7)	26 (53.1)	11 (24.4)
ADHF	19 (18.8)	9 (18.4)	14 (31.1)
Others	18 (17.8)	5 (10.2)	7 (15.6)
Palliative care team visit during STMCS, n, median (range)	n/a	3 (0–21)	3 (0–34)	0.97
Palliative care team visits after STMCS, n, median (range)	n/a	3 (0–10)	6 (0–49)	0.24
Inpatient death (%)	46 (45.5)	33 (67.3)	23 (51.1)	0.042
Palliative withdrawal of STMCS, n (%)	33 (32.7)	22 (44.8)	11 (24.4)	0.1

ADHF, acute decompensated heart failure; AMI, acute myocardial infarction; BMI, body mass index; n/a, not applicable; PCC, palliative care consultation; PCS, postcardiotomy shock; PMH, past medical history; STMCS, short-term mechanical circulatory support.

Mortality was significantly higher (p = 0.042) in the early PCC group (33 patients, 67.3%) than in the no PCC group (46 patients, 45.5%) or the late PCC group (23 patients, 51.1%). Mortality based on indications for STMCS was not significantly different (PCS, 43.6%; AMI, 54.4%; ADHF, 59.5%; others 53.3%. p = 0.446). Palliative withdrawal of STMCS was more frequent in the early PCC group (22 patients, 44.8%) than in the no PCC group (33 patients, 32.7%) or the late PCC group (11 patients, 24.4%), but the difference was not statistically significant.

Duration of STMCS, length of ICU stay after STMCS, and length of hospital stay after STMCS were all significantly shorter in the no PCC group compared with the early PCC group (Table 2). Duration of STMCS was similar between the early and late PCC groups, but ICU stay after STMCS and hospital stay after STMCS were significantly shorter in the early PCC group than in the late PCC group. Median time from initial PCC to hospital discharge was longer in the late PCC group (31 days) than in the early PCC group (18 days), but not significantly (p = 0.13).

Table 2.

Outcomes for All Patients

	No PCC median (range)	Early PCC median (range)	Late PCC median (range)	No PCC vs. early PCC	Early PCC vs. late PCC
	n = 101	n = 49	n = 45	p	p
Days on STMCS	4 (0–22)	12 (0–97)	18 (3–130)	<0.01	0.059
Days in the ICU after STMCS	11 (0–95)	19 (0–97)	38 (3–275)	0.004	<0.001
Days in hospital after STMCS	16 (0–134)	19 (0–156)	49 (14–275)	0.031	<0.001
Days from PCC to discharge	n/a	18 (0–152)	31 (3–267)	n/a	0.13

ICU, intensive care unit; n/a, not applicable.

Outcomes for deceased patients are summarized in Table 3. Duration of STMCS, ICU stay after STMCS, and hospital stay after STMCS were significantly shorter in the no PCC group than in the early PCC group; medians were all two days or less in the no PCC group. ICU stay and hospital stay after STMCS were significantly shorter in the early PCC group than in the late PCC group, but duration of STMCS did not differ. Median time from initial PCC to discharge (death) was similar (p = 0.537) between the early PCC group (seven days) and the late PCC group (eight days).

Table 3.

Outcomes for Deceased Patients

	No PCC median (range)	Early PCC median (range)	Late PCC median (range)	No PCC vs. early PCC	Early PCC vs. late PCC
	n = 46	n = 33	n = 23	p	p
Days on STMCS	1.5 (0–21)	9 (0–97)	17 (3–49)	<0.001	0.113
Days in the ICU after STMCS	2 (0–95)	11 (0–97)	26 (14–275)	<0.001	0.001
Days in hospital after STMCS	2 (0–95)	11 (0–97)	26 (14–275)	<0.001	0.001
Days from PCC to death	n/a	7 (0–102)	8 (3–267)	n/a	0.537

n/a, not applicable.

Discussion

In this retrospective cohort study, we examined the frequency of PCC in patients with cardiogenic shock who required STMCS and examined the relationship between the timing of PCC and patient outcomes. As far as we know, this is the first study to review PCC among ICU patients receiving STMCS. Because of high mortality in this population, previous research has focused on “hard” endpoints such as mortality, but it is also important to consider quality of life for patients and their families.¹⁷ Palliative care focuses on psychological, social, and spiritual aspects of care, and our data help provide a baseline for further study about the impact of palliative care on patients undergoing critical treatments such as STMCS.

Our most notable finding was that PCC was utilized in 48.2% of cases. The number of palliative care team visits during and after STMCS showed that, regardless of the timing of PCC, palliative care team remained involved in the care of patients throughout the clinical course, even after STMCS became not necessary. The frequency of PCC utilization in ICUs is not well documented. One study using a national database estimated that ∼14–20% of ICU admissions warrant PCC.¹⁸ A recently published study from our institution suggested that the frequency of PCC varies between different types of ICUs: 7.8% in cardiac care units versus 2.0% in cardiothoracic ICUs.¹⁹ Considering those findings, our data suggest that provision of PCC is relatively well established for patients with cardiogenic shock requiring STMCS at our institution. Furthermore, the number of visits during and after STMCS showed that, regardless of the timing of PCC, the palliative care team remained involved in the care of patients throughout the clinical course, even after STMCS was no longer necessary. This highlights the importance that palliative care teams can remain involved in survivorship, which can aid in both demonstrating the scope of palliative care practice, as well as alleviate concerns among providers that palliative care is congruent to end of life.

A second important finding was that, although overall time from initial PCC to discharge tends to be longer in the late PCC group, in deceased patients, initial PCC to death was similar between the early and late PCC. In addition, compared with the late PCC patients, the early PCC patients have higher mortality, shorter STMCS days, and higher rate of withdrawal of STMCS. These findings might suggest that PCC tends to occur toward the end of life regardless of the duration of STMCS; in the early PCC group, patients were sicker than the late PCC group and, therefore, received PCC earlier. In the late PCC group, PCC did not happen until patients remained on STMCS longer and the death became more certain.

Thirdly, the no PCC group had the lower mortality and the shorter time spent on STMCS, in the ICU and in the hospital. This demonstrates a dichotomous tendency in the no PCC group. If the patients died, it happened relatively shortly in two days, which might have been insufficient time to implement PCC. Otherwise, these patients showed they recovered rather quickly with shorter time on STMCS, in the ICU, and in the hospital. This might have made providers think PCC is less needed. Multiple barriers for incorporating PCC in ICUs have been pointed out in the literature, including denial of death (by clinicians, patient, or family), prognostic uncertainty, and “silos” of medical specialties among the heath care team.²⁰ The ideal timing for PCC might vary among institutions depending on the number of STMCS cases and the structure of the palliative care team. However, cardiogenic shock requiring STMCS is a condition in which only half of patients tend to survive over the course of one month. What patients and their families experience over that period can be significant as studies indicate significant levels of both depressive and post-traumatic stress symptoms in patients and caregivers up to one-year post-ICU^21,22; palliative care should be provided to alleviate their suffering in their most vulnerable moments.

Of note, indication for STMCS suggested that PCS was more prevalent in the late PCC and no PCC groups. Our survival rates for PCS compare favorably with reported outcomes in other studies.²³ However, it has been suggested that an attending surgeon had fewer PCC and fewer discussions of prognosis in the first 72 hours.²⁴ In addition to the barriers discussed earlier, surgical patients and surgical practice may present unique challenges for integration of palliative care into the ICU, because of poor knowledge and stigmas about palliative care.^25–27 This could have been one of the reasons of relatively late or no PCC in our cohort.

Limitations

There are several limitations to this study. This study took place in a single medical center, so the findings might not be generalizable to other institutions. Because it is a retrospective cohort study, we were unable to determine causality between PCC and outcomes. In addition, we lacked data about patients' clinical condition when PCC was requested; therefore, we could only speculate on the reasons that a PCC did not occur.

Future directions

There is no data about the clinical outcomes on patients on STMCS associated with PCC, including how PCC might impact survival, length of ICU or hospital stay, as well as on family's experience of anxiety, depression, post-traumatic stress, complicated grief, and care satisfaction. Our data showing higher rates of mortality and palliative withdrawal of STMCS in early PCC may be related to PCC assistance with aligning a patient and family's goals of care sooner than late PCC, and warrants additional exploration. Further studies are needed to evaluate the timing of PCC and impact on those clinical outcomes.

Conclusion

In conclusion, this retrospective cohort study showed that PCC was utilized for almost half of patients with cardiogenic shock requiring STMCS at our institution. PCC tends to occur toward the end of life regardless of the duration of STMCS. However, optimal timing for PCC remains unclear. Cardiogenic shock requiring STMCS is a condition with high in-hospital mortality. Further studies are needed to evaluate timing of PCC and clinical outcomes associated with PCC in this population.

Footnotes

Author Disclosure Statement

Dr. Naka is a consultant for Abbott. All other authors have no conflicts of interest to disclose.

References

Thiele

, Zeymer

, Neumann

F-J

, et al.: Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med, 2012; 367:1287–1296.

Waha

, Desch

, Eitel

, et al.: What is the evidence for IABP in STEMI with and without cardiogenic shock?. Ther Adv Cardiovasc Dis, 2012; 6:123–132.

Garan

, Eckhardt

, Takeda

, et al.: Predictors of survival and ability to wean from short-term mechanical circulatory support device following acute myocardial infarction complicated by cardiogenic shock. Eur Heart J Acute Cardiovasc Care, 2017:2048872617740834 [Epub ahead of print]; DOI: 10.1177/2048872617740834.

Stretch

, Sauer

, Yuh

, Bonde

: National trends in the utilization of short-term mechanical circulatory support: Incidence, outcomes, and cost analysis. J Am Coll Cardiol, 2014; 64:1407–1415.

Kelley

, Morrison

: Palliative care for the seriously ill. N Engl J Med, 2015; 373:747–755.

Campbell

, Guzman

: Impact of a proactive approach to improve end-of-life care in a medical ICU. Chest, 2003; 123:266–271.

Campbell

, Guzman

: A proactive approach to improve end-of-life care in a medical intensive care unit for patients with terminal dementia. Crit Care Med, 2004; 32:1839–1843.

Norton

, Hogan

, Holloway

, et al.: Proactive palliative care in the medical intensive care unit: Effects on length of stay for selected high-risk patients. Crit Care Med, 2007; 35:1530–1535.

Chanques

, Jaber

, Barbotte

, et al.: Impact of systematic evaluation of pain and agitation in an intensive care unit. Crit Care Med, 2006; 34:1691–1699.

10.

Schneiderman

, Gilmer

, Teetzel

, et al.: Effect of ethics consultations on nonbeneficial life-sustaining treatments in the intensive care setting: A randomized controlled trial. JAMA, 2003; 290:1166–1172.

11.

O'Mahony

, McHenry

, Blank

, et al.: Preliminary report of the integration of a palliative care team into an intensive care unit. Palliat Med, 2010; 24:154–165.

12.

Penrod

, Deb

, Luhrs

, et al.: Cost and utilization outcomes of patients receiving hospital-based palliative care consultation. J Palliat Med, 2006; 9:855–860.

13.

Morrison

, Penrod

, Cassel

, et al.: Cost savings associated with US hospital palliative care consultation programs. Arch Intern Med, 2008; 168:1783–1790.

14.

Kaufer

, Murphy

, Barker

, Mosenthal

: Family satisfaction following the death of a loved one in an inner city MICU. Am J Hosp Palliat Care, 2008; 25:318–325.

15.

Lilly

, De Meo

, Sonna

, et al.: An intensive communication intervention for the critically ill. Am J Med, 2000; 109:469–475.

16.

Lautrette

, Darmon

, Megarbane

, et al.: A communication strategy and brochure for relatives of patients dying in the ICU. N Engl J Med, 2007; 356:469–478.

17.

Allen

, Stevenson

, Grady

, et al.: Decision making in advanced heart failure: A scientific statement from the American Heart Association. Circulation, 2012; 125:1928–1952.

18.

Hua

, Li

, Blinderman

, Wunsch

.: Estimates of the need for palliative care consultation across united states intensive care units using a trigger-based model. Am J Respir Crit Care Med, 2014; 189:428–436.

19.

Stix

, Wunsch

, Clancy

, et al.: Variability in frequency of consultation and needs assessed by palliative care services across multiple specialty ICUs. Intensive Care Med, 2016; 42:2104–2105.

20.

Nelson

: Identifying and overcoming the barriers to high-quality palliative care in the intensive care unit. Crit Care Med, 2006; 34(11 Suppl):S324–S331.

21.

Cameron

, Chu

, Matte

, et al.: One-year outcomes in caregivers of critically ill patients. N Engl J Med, 2016; 374:1831–1841.

22.

Parker

, Sricharoenchai

, Raparla

, et al.: Posttraumatic stress disorder in critical illness survivors: A metaanalysis. Crit Care Med, 2015; 43:1121–1129.

23.

Flécher

, Anselmi

, Corbineau

, et al.: Current aspects of extracorporeal membrane oxygenation in a tertiary referral centre: Determinants of survival at follow-up. Eur J Cardiothorac Surg, 2014; 46:665–671.

24.

Kross

, Engelberg

, Downey

, et al.: Differences in end-of-life care in the ICU across patients cared for by medicine, surgery, neurology, and neurosurgery physicians. Chest, 2014; 145:313–321.

25.

Aslakson

, Wyskiel

, Thornton

, et al.: Nurse-perceived barriers to effective communication regarding prognosis and optimal end-of-life care for surgical ICU patients: A qualitative exploration. J Palliat Med, 2012; 15:910–915.

26.

Katz

: The term “supportive care” is preferable to “palliative care” for consults in the cardiothoracic intensive care unit. J Thorac Cardiovasc Surg, 2018; 155:2030–2031.

27.

Nakagawa

, Hua

, Takayama

: Who is not comfortable with the term “palliative care”—patient, family, or surgeon?. J Thorac Cardiovasc Surg, 2018; 155:2032–2033.