- Email: [email protected]
Outcomes After Intracranial Hemorrhage in Patients with Left Ventricular Assist Devices: A Systematic Review of Literature
Literature Review
Outcomes After Intracranial Hemorrhage in Patients with Left Ventricular Assist Devices: A Systematic Review of Literature Theresa Elder1, Alankrita Raghavan4, Arvin Smith4, Christina Huang Wright1,4, James Wright1,4, Christopher Burant5,6, Martha Sajatovic3,2, Alan Hoffer1,4
Key words Hemorrhagic stroke - ICH - Intracranial hemorrhage - LVAD - Ventricular assist device -
Abbreviations and Acronyms ICH: Intracranial hemorrhage INR: International normalized ratio LVAD: Left ventricular assist device From the Departments of 1Neurological Surgery, 2Neurology, and Psychiatry and 3Neurological and Behavioral Outcomes Research Center, University Hospitals Cleveland Medical Center, Cleveland, Ohio; and 4School of Medicine, 5Mandel School of Applied Social Sciences, and 6Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, Ohio, USA To whom correspondence should be addressed: James Wright, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019) 132:265-272. https://doi.org/10.1016/j.wneu.2019.08.211 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
INTRODUCTION Intracranial hemorrhages (ICHs) are common neurosurgical complications associated with high mortality.1-3 Known causes of ICH include hypertensive arteriosclerosis and arteriovenous abnormalities, as well as brain tumors. Patients with left ventricular assist devices (LVADs) are at increased risk of ICH in particular, because of a requirement for continuous anticoagulation to prevent device thrombosis and ischemic stroke.4 The incidence of ICH in patients with LVAD has been reported to fall between 3% and 11% in multiple outcomes studies.5-7 Although LVADs are known to improve survival in patients with advanced heart failure,8 the resultant changes in systemic blood flow patterns cause additional bleeding risks irrespective of anticoagulation requirements, such as
- BACKGROUND:
Intracranial hemorrhage (ICH) is a common complication in patients with left ventricular assist devices (LVADs) and carries a dismal prognosis. Limited data are available on the management and outcomes in this population, which are essential to determine an optimal treatment strategy. We conducted a systematic literature review to determine the clinical characteristics and survival of this population.
- METHODS:
PubMed, Ovid, Embase. Scopus, Cochrane, CENTRAL, and Web of Science articles were selected using the following terms: (“Heart-Assist Devices” or “left ventricular assist device” or “LVAD”) AND (“intracranial hemorrhage” OR “cerebral hemorrhage” OR “brain hemorrhage” OR “intracerebral hemorrhage” OR “intraparenchymal hemorrhage” OR “hemorrhagic stroke”). Abstracts and articles were screened according to inclusion and exclusion criteria that were determined a priori. Potential studies were reviewed by 4 authors, who reached a consensus on the final studies to be included.
- RESULTS:
The literature review yielded 609 abstracts, which were screened according to predetermined inclusion criteria. A total of 143 full-text articles were reviewed, and 8 articles were included in the final qualitative analysis. These studies reviewed data for 597 patients with LVADs who had ICH. The mortality for ICH was widely variable across studies and ranged from 16% to 100%.
- CONCLUSIONS:
There is minimal existing literature on patients with LVAD with ICH that report patient outcomes in a nonstandardized fashion. The studies included in this analysis report mortality consistent with previous reports, indicating a need for further investigation to identify risk factors and improve outcomes in these patients.
acquired von Willebrand disease.9 As a result, patients on LVADs are at increased risk for ICH.10 When LVADs were introduced, they were commonly used as a bridge to transplantation. However, with ongoing donor shortages in combination with advancing technology, an increased proportion of patients are using LVADs as destination therapy.6,11 More than 26,000 patients have received LVAD implants in the last decade, and the number of LVAD implants has shown an average annual increase of 12.6%.12,13 Furthermore, the proportion of patients using LVADs as destination therapy increased from 14.6% in 2007 to 41.6% in 2011, with a recent
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
study citing the average duration of LVAD therapy to be 581 days.14,15 These patients remain on long-term anticoagulation and antiplatelet regimens for longer periods, putting them at further risk for spontaneous bleeding. Previous reports have suggested that as many as 50% of patients on LVADs have bleeding incidents within 30 days of device implantation.16 Gastrointestinal bleeding in patients with LVADs has been well documented in the literature, and with extensive discussion of its management.17 However, ICH risk factors and management strategies have received less attention, with their sole mention in single-institution studies and case series.
www.journals.elsevier.com/world-neurosurgery
265
LITERATURE REVIEW THERESA ELDER ET AL.
Furthermore, these studies often combine patients with ischemic and hemorrhagic stroke, with no distinction in management strategies between the 2 diseases. With as many as two thirds of documented ICH cases resulting in patient death, further investigation is warranted to better understand and treat this complication.4 The goal of our systematic literature review is to understand survival and mortality of ICH in patients with LVADs, specifically investigating the association between anticoagulation regimens and device type with ICH outcome. These data may aid in establishing potential ICH risk assessments for patients with LVADs and help determine optimal treatment strategies for ICH in this patient population. METHODS A systematic review of the reported data was performed in accordance with the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analyses) guidelines. PubMed, Ovid, Embase. Scopus, Cochrane, CENTRAL, and Web of Science articles were selected using the following terms: (“Heart-Assist Devices” or “left ventricular assist device” or “LVAD”) AND (“intracranial hemorrhage” OR “cerebral hemorrhage” OR “brain hemorrhage” OR “intracerebral hemorrhage” OR “intraparenchymal hemorrhage” OR “hemorrhagic stroke”). The reference lists of selected articles were screened for additional articles. Potential studies were reviewed by 4 authors (A.R., A.S., J.W., and C.W.), who reached a consensus on the studies to be included in the present review. The reference lists of all selected reports were screened for additional studies. The article inclusion and exclusion criteria were defined a priori. The inclusion criteria were as follows: original research reports regarding patients with mechanical assist devices who have ICH, regarding data on mortality, morbidity, and discharge deposition, regarding bleeding risk and outcomes with mention of ICH, case series, retrospective and prospective studies, randomized controlled trials, systematic reviews, and clinical trials. The exclusion criteria were as follows: articles regarding only ischemic stroke, case reports on <5 patients, technical notes, letters, review articles, articles in
266
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
languages other than English, animal trials, and in vitro studies. Articles were excluded if they did not address the outcomes of interest (i.e., mortality and survival). Studies were also excluded if the mortality and survival were not clear and/ or they did not explicitly indicate results. All included studies were evaluated by the authors for the quality of evidence provided. Using criteria outlined by Balshem et al.,18 each article was rated into 1 of 4 levels of quality (Table 1). Factors that increased the rating of a study included the size of the study effect, evidence of a dose-response gradient, and lack of confounding variables. Ratings were decreased for inconsistency, indirectness, imprecision, and risk of bias. This rating system was chosen based on its ability to assess an entire body of evidence rather than strictly the evidence of an individual study and helped determine the strength of conclusions drawn. This study was a systematic review and was therefore exempt from institutional review board approval. RESULTS The search yielded 609 abstracts, 143 of which were selected for full-text screening. Several studies were excluded because of failure to provide appropriately calculated mortality or survival metrics. Eight articles were selected for inclusion in the review (Figure 1). All articles published before August 2018 were included in the search. Six studies were retrospective whereas 2 were prospective. All studies were from the United States with the exception of that of Muthiah et al.,25 which was from Australia. Two studies reported external funding: Yavar et al.19 received research funds from Abbot and Medtronic and Muthiah et al.25 received funding from the Pfizer Competitive Cardiovascular Lipid Research Grant and the St. Vincent’s Clinic Grant. The data examined were collected over 14 years (2003e2017), with publication years spanning 2013e2018 (Table 1). These studies reviewed data for of 597 patients with LVADs who had ICH. Acharya et al.20 had the largest sample, describing 423 patients with ICH from Inter-agency Registry for Mechanically Assisted Circulatory Support
(INTERMACS) who received continuousflow LVADs. The mean age of patients studied across all cohorts was 55 years, although 2 of these studies reported the mean age of patients with ICH only. There was a male predominance in the entire population of patients with LVAD as well as in those with ICH, with males comprising 78% of total patients and 64% of the patients with ICH. The incidence of ICH among female patients with LVAD was twice as high as in males, at 12.32% and 6.13%, respectively. Seven studies examined continuous-flow LVADs, whereas one examined both pulsatile and continuous-flow devices. The most frequently implanted device was the Heartmate II (Abbott, Chicago, Illinois, USA), followed by HeartWare (HeartWare International Inc., Framingham, Massachusetts, USA), with no studies including patients with Heartmate III (Abbott) (Table 2). Hemorrhage Characteristics and Analysis of Risk Factors Four studies included the ICH hemorrhage pattern in their data, with intraparenchymal hemorrhage representing most hemorrhages (42%e61%).10,21,22,24 Subarachnoid hemorrhage comprised 24%e48% of all hemorrhages, and both subdural and epidural hemorrhages combined accounted for the remaining 0%e 10% of hemorrhages. Average international normalized ratio (INR) ranged from 1.3 to 3.8 among patients with ICH, with a mean of 2.7 across all studies. All patients were on an anticoagulation regimen at the time of ICH, most of which included warfarin, with the exception of 9 patients,10,22 who were taking either heparin or an alternative nonwarfarin anticoagulant. Six studies reported antiplatelet regimens, all which included aspirin with or without an additional agent, with the exception of 1 patient in Ramey et al.‘s study.10 The mean duration of LVAD therapy before hemorrhage was 454 days. Management and Outcomes Analysis Four studies included the percentage of patients who underwent surgical intervention, which ranged from 0% to 40%.10,21-23 The time points at which mortality and survival were reported varied from same admission mortality to 2-year mortality. Mortality from ICH was widely
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
LITERATURE REVIEW
Low No cohorts, 1 group of patients with LVAD
DISCUSSION
Single Prospective
LVAD, left ventricular assist device.
Australia 2011e2012 Journal of Heart and Lung Transplantation Muthiah et al., 201625
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
varied across studies and ranged from 16% to 100%; however, the 100% mortality was drawn from a population of 2 patients.23 Of the 3 studies reporting 30-day survival, data ranged from 45.3% to 67%, whereas data reported for 2-year survival ranged from 0% to 61% (Table 3).19-21,23
28
Moderate 4-factor prothrombin complex concentrate assisted reversal, reversal with traditional agents, no reversal 2008e2015 ASAIO Wong et al., 201624
United States
Prospective cohort
Single
237
Very low One group of patients with LVAD 2006e2014 ASAIO Chou et al., 201723
United States
Retrospective case series
Single
9
Moderate One group of patients with LVAD with intracranial hemorrhage 2007e2016 World Neurosurgery Tahir et al., 201822
United States
Retrospective cohort
Multicenter
59
Moderate No cohorts, 1 group of patients with LVAD 2003e2012 Journal of Neurosurgery Wilson et al., 201321
United States
Retrospective cohort
Single
330
Low No cohorts, 1 group of patients with LVAD 2012e2015 JACC: Heart Failure Acharya et al., 201720
United States
Retrospective database
Multicenter
7112
Low Any bleeding event, no bleeding event 2008e2017 ASAIO Yavar et al., 201719
United States
Retrospective cohort
Single
375
Moderate Total artificial heart or LVAD 2013e2016 World Neurosurgery Ramey et al., 201710
United States
Retrospective cohort
Single
58
Quality of Evidence Total Participants in Study Number of Centers Study Type Country Years Investigated Journal Title Reference
Table 1. Studies Included in Review
ICH IN PATIENTS WITH LVAD
Cohorts
THERESA ELDER ET AL.
This systematic literature review surveys the available literature on survival and mortality in patients on LVADs with concomitant ICH. As the number of advanced heart failure patients using LVADs continues to increase, more patients will be at risk for ICH as a complication of therapy. Although ICH in the general population is known to carry a poor prognosis, survival data in patients on LVADs are limited and have been reported in a nonstandardized fashion across studies, making risk analysis and outcome determination challenging. ICH is a risk associated with any disease process requiring long-term anticoagulation. However, the risk is disproportionately higher in patients with LVAD compared with other patient populations requiring anticoagulants, such as atrial fibrillation.26 Patients with atrial fibrillation require anticoagulation to target an INR range equal to that of patients with LVAD but have a significantly lower ICH rate, ranging between 0.3% and 0.6% per year.16 A trial by Granger et al.27 compared the effects of various anticoagulants used in atrial fibrillation and cited a 0.47% rate of ICH per year for patients taking warfarin. This trial reported an increase in ICH rate by approximately 40% when aspirin was added to the anticoagulation regimen. Although anticoagulation therapy does play an important role in the increased rate of bleeding complications in the population with LVAD, it is also believed to relate to changes in hemostatic factors causing an acquired von Willebrand disease state.4,28 The nonphysiologic pump hemodynamics place shear stress on circulating blood components, resulting in impaired platelet aggregation similar to that seen in aortic stenosis, but which is reversible on removal of the device.16,29 Continuous-flow LVADs were designed in effort to reduce blood flow pulsatility,
www.journals.elsevier.com/world-neurosurgery
267
LITERATURE REVIEW THERESA ELDER ET AL.
which resulted in an increase in 2-year survival in patients with heart failure using these devices compared with those with pulsatile-flow LVADs.5,7 However, continuous-flow LVADs have been shown to have an increased rate of bleeding complications, with the ICH rate being 8% and 11% for pulsatile-flow and continuous-flow LVADs, respectively.6 Ramey et al.10 noted that 100% of their patient population tested positive for acquired von Willebrand disease, as diagnosed by decreased level of von Willebrand factor multimers, and subsequently emphasized the importance of antiplatelet agent reversal. In another study by Uriel et al.,9 all 31 patients with continuous-flow LVADs were found to have acquired von Willebrand disease, and 58% of these patients had bleeding events. Risk Factors Multiple variables have been investigated as possible risk factors for ICH. Studies stratifying patient outcomes based on hemorrhage pattern consistently reported intraparenchymal hemorrhages to be the most common and also to carry the worst prognosis.10,21,22,24 In Wilson et al.‘s series,21 all 5 ICH cases that necessitated surgical intervention were spontaneous intraparenchymal bleeds. However, the survival at 60 days for these patients was only 20%, bringing into question the survival benefit of operative intervention. Furthermore, most of the reported hemorrhages were spontaneous and carried a worse prognosis compared with traumatic ICH, with the mortality of traumatic subarachnoid hemorrhage reported at 0% in 1 cohort.10,21,22 This trend may be related to the described nonphysiologic hemodynamics caused by the rotary mechanism of the LVADs, leading to platelet dysfunction and increased bleeding risk.28 Another risk factor that emerged in our review was the impact of gender on ICH incidence and mortality. We found that although females composed only 20% of the total patients with LVAD included in this review, they accounted for 36% of the patients with ICH. Yavar et al.19 reported that female patients had a 60% higher risk of hemorrhage compared with male patients, and 5 studies included in this review reflected similar findings, with females having approximately 50%
268
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
increased risk of hemorrhage compared with male patients.19-23 The mechanism behind this characteristic is likely multifactorial. One hypothesis relates to the presence of gender effects on platelet function caused by hormone-related mechanisms.19 There is a known protective effect of endogenous estrogen on stroke rates, because premenopausal women sustain fewer strokes than do age-matched men, whereas postmenopausal women sustain more strokes than do age-matched men.30,31 It is also known that oral contraceptives increase the risk for stroke via hormone-mediated upregulation of endothelial dysfunction.32,33 Considering that most women represented in our review were of postmenopausal age, there is reason to believe that hormone signaling leads to a greater upregulation of LVAD-mediated endothelial dysfunction in female patients than it does in male patients. Although other hypotheses exist that put females at greater risk of bleeding complications
from LVADs, such as decreased body surface area and increased frailty, the exact mechanism remains unclear.33 Two studies in our analysis did not report a difference in hemorrhage risk between genders: Chou et al.23 reported a 1:1 female/male ratio, although their sample size was limited to only 2 patients, whereas Wong et al.24 did not provide a gender breakdown of ICH incidence or mortality. However, even among the studies that did report on gender, there was no significant effect of gender on mortality after ICH. Tahir et al.‘s22 was the largest study to reflect this situation, reporting a total cohort comprised of 32% females, who accounted for 28% of deaths. One commonly reported parameter that was associated with increased mortality was Glasgow Coma Scale score.21 Tahir et al.22 also found that both the extent of midline shift and the size of ICH on initial presentation were associated with poorer outcomes. In that study, patients
Figure 1. PRISMA systematic review flow diagram.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
LITERATURE REVIEW THERESA ELDER ET AL.
ICH IN PATIENTS WITH LVAD
Table 2. Description of Studies
Reference
Mean Age of Entire Study Cohort (years)
Sex (Male/Female) of Entire Study Cohort
Sex (Male/Female) of Patients with ICH Only
Race
Type of LVAD
Ramey et al., 201710
56.7 (patients with ICH only)
49:9
5:0
—
49/58 LVAD (type unspecified), 10/58 TAH, 1/58 both LVAD þ TAH
—
Yavar et al., 201719
55.6 0.7
291:94
24:13
White: 82% Other: 18%
100% continuous
—
Acharya et al., 201720
Not specified
5579:1533
228:135
—
100% continuous
—
Wilson et al., 201321
54 12
253:77
67:33
White: 78%, black: 18%, Asian: 1%, Hispanic: 1%
30% continuous, 70% pulsatile (patients with ICH: 42% continuous, 58% pulsatile)
—
Tahir et al., 201822
54.3 (patients with ICH only)
40:19
40:19
For patients with ICH: white, 54.2%; black, 45.8%
72.2% Heartmate II and continuous, 14.8% Heartware HD and pulsatile, 13% not specified
39 Heartmate II 8 Heartware 7 other
Chou et al., 201723
56.5 12
7:2
1:1
—
100% continuous
6 Heartmate II 3 Heartware
Wong et al., 201624
52 13
—
18:13
—
100% continuous
100% Heartmate II
Muthiah et al., 201625
48 14
25:3
—
—
100% continuous
100% HeartWare
LVAD Brand Name
ICH, intracranial hemorrhage; LVAD, left ventricular assist device; TAH, total artificial heart.
with any amount of midline shift were found to be 35% less likely to survive compared with patients with no shift on initial head computed tomography. Management and Outcomes Anticoagulation and antiplatelet regimens were inconsistently reported across the analyzed studies. In the study by Wilson et al.,21 all patients were taking aspirin and warfarin, which was the suggested anticoagulation regimen in the recent Heartmate II Pivotal Trial, with an INR target range of 2e3. Most studies report a lack of standardized anticoagulation regimen used in their study populations and provide insufficient detail on the regimens used. For example, the study by Uriel et al.9 reported that 68.3% of patients were taking warfarin, 55.7% were taking aspirin, and 58.2% were taking dipyridamole but did not specify which patients received dual-agent therapy, which could significantly affect outcomes. Although long-term anticoagulation regimens varied across the
studies included in this review, those that included INR values in their data did not show a correlation between INR and risk of ICH development. INR goals for patients with LVAD, regardless of the pump type, are generally in the range of 2.0e 3.0.34 Tahir et al.22 found that 37% of their patients with ICH were subtherapeutic with INR <2.0, and Ramey et al.10 found that only 20% of patients with ICH had a supratherapeutic INR. Furthermore, although ICH occurred independently of INR, the anticoagulation reversal strategy used often varied based on INR value at presentation, adding an additional variable that could affect outcome. For example, 66% of patients in the series reported by Tahir et al.22 underwent anticoagulation reversal, whereas in the series reported by Ramey et al.,10 40% of patients received anticoagulation reversal, 40% discontinued anticoagulation, and 20% had no intervention. Use of reversal agents was nonuniform across the included studies. Wilson et al.21
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
consistently reversed antiplatelet agents and warfarin with administration of platelet transfusions and fresh frozen plasma, respectively, whereas Ramey et al.10 used an unspecified combination of fresh frozen plasma, prothrombin complex concentrate, and/or vitamin K with a goal INR <1.4. Tahir et al.22 did not specify their INR reversal strategy but did mention that patients taking novel anticoagulants such as rivaroxoban or dabigatran posed a challenge to rapid reversal. This report was published in 2018, 5 years after Wilson et al.‘s study, highlighting the impact of recent pharmacologic advancements on this clinical scenario. Although most studies recommend reversal or temporary discontinuation of anticoagulation as part of ICH management, the concern for subsequent pump thrombosis and other thrombotic events leads to uncertainty regarding the duration of anticoagulant discontinuation. Even although ICH rate may not correlate with increasing INR, the risk of thrombosis is
www.journals.elsevier.com/world-neurosurgery
269
www.SCIENCEDIRECT.com
Reference
Total Number of Study Incidence Participants of ICH (n/N)
Ramey et al., 201710
58
Hemorrhage Location (%)
Mortality, as Reported (%)
Survival, as Reported (%)
Anticoagulation Regimen (%)
Antiplatelet Regimen (%)
5
Subarachnoid: 40 Intraparenchymal: 60 Epidural/subdural: 0
60
40
80 on warfarin, 20 on heparin
80 on aspirin with or without dipyridamole
Mean Mean Duration International of Left Ventricular Normalized Ratio of % with Assist Device Surgical Before Patients with ICH Intervention Hemorrhage 40
32 months (range, 2e84 months)
2.6
37
—
—
38 at 2 years
100 on warfarin
100 on aspirin
—
158 days
—
7112
423
—
70 at 1 year
45.3 at 1 month 34.8 at 6 months 30.3 at 1 year
—
—
—
—
—
Wilson et al., 201321
330
36
Subarachnoid: 28 33 at 30 days, Intraparenchymal: 47 39 at 1 year Epidural/subdural: 22
45 at 4 years
100 on warfarin
100 on aspirin
14
300 days
—
Tahir et al., 201822
598
59
Subarachnoid: 24 Intraparenchymal: 61 Epidural/subdural: 5
42.4 same admission
—
8.5
470 days
2.85
Chou et al., 201723
9
2
—
100
50 at discharge, 0 at 1 year
1/2 warfarin 1/2 heparin
100 aspirin
0
122 days
1.3
Wong et al., 201624
237
42
Subarachnoid: 48 Intraparenchymal: 42 Epidural/subdural: 10
16
—
—
—
—
578 (445) days
—
Wong et al., 201624 (first cohort)
237
10
—
20
—
—
—
—
661 days
3.8
Wong et al., 201624 (second cohort)
237
10
—
20
—
—
—
—
453 days
3.5
Wong et al., 201624 (third cohort)
237
11
—
9.1
—
—
—
—
618 days
2.4
Muthiah et al., 201625
28
3
—
66.7
—
100 warfarin
100 aspirin and clopidogrel
—
223 days
—
ICH, intracranial hemorrhage.
95 on anticoagulation 95 on unspecified regimen; antiplatelet regimen 88 coumadin, 12 noncoumadin
LITERATURE REVIEW
375
Acharya et al., 201720
ICH IN PATIENTS WITH LVAD
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
Yavar et al., 201719
THERESA ELDER ET AL.
270 Table 3. Mortality and Intracranial Hemorrhage Data
LITERATURE REVIEW THERESA ELDER ET AL.
inversely related to INR. Nassif et al.35 reported data on 11,000 INRs drawn from patients with LVAD and found that the highest thrombosis rate occurred in patients with INR <1.5, and high rates were also seen with INRs in the 1.5e1.99 range. The rate of thrombotic events was also reported as a function of INR, citing a hazard ratio of 0.39 (95% confidence interval, 0.22e0.70; P ¼ 0.002). When balancing the risk of thrombosis with the risk of bleeding, this study determined the ideal INR goal to be 2.6.35 Ironically, this figure is identical to the average INR seen among the patients with ICH in the study by Ramey et al.,10 reflective of this challenging clinical dilemma. There is no clear understanding of the risk for thrombotic events when aspirin and plavix are withheld or reversed. Wong et al.24 studied the effects of withholding anticoagulation in patients with LVAD with ICH and reported that 11 of their cohort of 31 patients were managed successfully without anticoagulation reversal. However, these patients collectively had a mean ICH volume of 0.4 mL, in contrast to Tahir et al.‘s22 cohort of 19 patients who did not undergo reversal of anticoagulation, who had an average ICH volume of 19 mL. Wilson et al.21 reported no thrombotic events in their study population after reversal of warfarin (with platelets given in 39% of patients and fresh frozen plasma in 61% of patients), with a median duration of suspension for aspirin and warfarin being 6 and 10.5 days, respectively, concluding that aspirin can safely be withheld for 1 week and warfarin can be safely withheld for 10 days. Ramey et al.10 suggested a stepwise algorithm for anticoagulation management, with criteria for reversal based on patients’ neurologic deficit and the size of initial hemorrhage on computed tomography. Per their protocol, only large ICHs (defined as intraparenchymal hemorrhage >3 cm or subarachnoid hemorrhage >1e2 mm thickness) required anticoagulation reversal, but the duration of discontinuation was variable. Furthermore, these recommendations were not applicable to patients in whom surgical intervention was required. The mortality reported in our review was widely variable, partially because of
ICH IN PATIENTS WITH LVAD
inconsistent data collection time points. Some studies defined survival as survival to discharge from the hospital, whereas others reported 1-year and 2-year survival.19-23 Although some studies further stratify survival by functional status, such as the Glasgow Outcome Score or the modified Rankin Score, others do not provide such data.21,22 The lack of standardized reporting challenges the ability to make associations between intervention strategies and outcomes. Future Directions We suggest multiple areas of focus for future studies to help determine an optimal therapeutic approach and improve outcomes. First, patients should be stratified according to the anticoagulation regimen used, specifically distinguishing patients on multiple therapeutic agents from those on single-agent therapy. Second, studies must follow a consistent protocol for anticoagulation reversal, with a specific target INR value as a result of the intervention. Furthermore, the timing of INR reversal should be specified relative to the time of ICH diagnosis, with repeat INR values reported and trended relative to the ICH progression. Outcomes and survival should be measured at consistent time points, ideally at 30 days, 90 days, and 1 year time points, because this will help identify and differentiate between early and late bleeding risks. Limitations The inconsistencies in data reporting contributed to several limitations in our review. Certain variables were reported to affect outcomes in some studies, but were not reported at all in others, such as dual antiplatelet therapy usage or anticoagulation reversal. This factor limited the quality of evidence reported in the studies reviewed as a result. The type of LVAD was often not specified, and Ramey et al.10 included 10 patients with a total artificial heart in their ICH cohort, which affects blood flow hemodynamics and, potentially, the pathophysiologic mechanism of ICH in those patients. Six of the 8 studies were retrospective, and only studies from the United States and Australia provided enough data to be included in our review. Three of the studies contained 5 patients with ICH. These small cohorts contributed to the
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
wide ranges reported in outcome measures, particularly the series of 2 patients reported by Chou et al.,23 which showed a 100% 60-day mortality. Although the use of LVADs as destination or longer bridge-to-transplant therapy may primarily be an issue of developed countries, this will become a global issue as more countries undergo the epidemiologic transition. Further data from developing countries are needed to identify regionspecific variables affecting survival. CONCLUSIONS ICH is a common complication in patients with LVAD and carries a grim prognosis. The mortality reported in the studies included in this analysis is consistent with historical reports of mortality in ICH, and further investigation is essential to gaining a better understanding of the prevention and treatment of ICH in patients with LVAD. Future studies should focus on elucidating the association between treatment and survival outcomes, as well as exploring the impact of various anticoagulation regimens on this devastating clinical scenario. REFERENCES 1. Fogelholm R, Murros K, Rissanen A, Avikainen S. Long term survival after primary intracerebral haemorrhage: a retrospective population based study. J Neurol Neurosurg Psychiatry. 2005;76: 1534-1538. 2. Flaherty ML, Haverbusch M, Sekar P, et al. Location and outcome of anticoagulant-associated intracerebral hemorrhage. Neurocrit Care. 2006;5: 197-201. 3. Sacco S, Marini C, Toni D, Olivieri L, Carolei A. Incidence and 10-year survival of intracerebral hemorrhage in a population-based registry. Stroke. 2009;40:394-399. 4. Cho SM, Moazami N, Frontera JA. Stroke and intracranial hemorrhage in HeartMate II and HeartWare left ventricular assist devices: a systematic review. Neurocrit Care. 2017;27:17-25. 5. Miller LW, Pagani FD, Russell SD, et al. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007;357:885-896. 6. Slaughter MS. Long-term continuous flow left ventricular assist device support and end-organ function: prospects for destination therapy. J Card Surg. 2010;25:490-494. 7. Pagani FD, Miller LW, Russell SD, et al. Extended mechanical circulatory support with a continuousflow rotary left ventricular assist device. J Am Coll Cardiol. 2009;54:312-321.
www.journals.elsevier.com/world-neurosurgery
271
LITERATURE REVIEW THERESA ELDER ET AL.
8. Rose EA, Gelijns AC, Moskowitz AJ, et al. Longterm use of a left ventricular assist device for endstage heart failure. N Engl J Med. 2001;345: 1435-1443. 9. Uriel N, Pak SW, Jorde UP, et al. Acquired von Willebrand syndrome after continuous-flow mechanical device support contributes to a high prevalence of bleeding during long-term support and at the time of transplantation. J Am Coll Cardiol. 2010;56:1207-1213. 10. Ramey WL, Basken RL, Walter CM, Khalpey Z, Lemole GM, Dumont TM. Intracranial hemorrhage in patients with durable mechanical circulatory support devices: institutional review and proposed treatment algorithm. World Neurosurg. 2017;108:826-835. 11. Birks EJ. A changing trend toward destination therapy: are we treating the same patients differently? Tex Heart Inst J. 2011;38:552-554. 12. Briasoulis A, Inampudi C, Akintoye E, Adegbala O, Alvarez P, Bhama J. Trends in utilization, mortality, major complications, and cost after left ventricular assist device implantation in the United States (2009 to 2014). Am J Cardiol. 2018; 121:1214-1218. 13. Patel N, Kalra R, Doshi R, Bajaj NS, Arora G, Arora P. Trends and cost of heart transplantation and left ventricular assist devices. JACC: Heart Fail. 2018;6:424. 14. Tsiouris A, Paone G, Nemeh HW, et al. Short and long term outcomes of 200 patients supported by continuous-flow left ventricular assist devices. World J Cardiol. 2015;7:792-800. 15. Health Quality Ontario. Left ventricular assist devices for destination therapy: a health technology assessment. Ont Health Technol Assess Ser. 2016;16: 1-60. 16. Klovaite J, Gustafsson F, Mortensen SA, Sander K, Nielsen LB. Severely impaired von Willebrand factor-dependent platelet aggregation in patients with a continuous-flow left ventricular assist device (HeartMate II). J Am Coll Cardiol. 2009;53: 2162-2167. 17. Cushing K, Kushnir V. Gastrointestinal bleeding following LVAD placement from top to bottom. Dig Dis Sci. 2016;61:1440-1447.
272
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
18. Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64:401-406. 19. Yavar Z, Cowger JA, Moainie SL, Salerno CT, Ravichandran AK. Bleeding complication rates are higher in females after continuous-flow left ventricular assist device implantation. ASAIO J. 2018; 64:748-753. 20. Acharya D, Loyaga-Rendon R, Morgan CJ, et al. INTERMACS analysis of stroke during support with continuous-flow left ventricular assist devices: risk factors and outcomes. JACC Heart Fail. 2017;5:703-711. 21. Wilson TJ, Stetler WR, Al-Holou WN, Sullivan SE, Fletcher JJ. Management of intracranial hemorrhage in patients with left ventricular assist devices. J Neurosurg. 2013;118:1063-1068. 22. Tahir RA, Rotman LE, Davis MC, et al. Intracranial hemorrhage in patients with a left ventricular assist device. World Neurosurg. 2018;113:e714-e721. 23. Chou J, Bermudez C, Kormos R, Teuteberg J. Permanent continuous flow left ventricular assist devices use after acute stabilization for cardiogenic shock in acute myocardial infarction. ASAIO J. 2017;63:e13-e17. 24. Wong JK, Chen PC, Falvey J, et al. Anticoagulation reversal strategies for left ventricular assist device patients presenting with acute intracranial hemorrhage. ASAIO J. 2016;62:552-557. 25. Muthiah K, Connor D, Ly K, et al. Longitudinal changes in hemostatic parameters and reduced pulsatility contribute to non-surgical bleeding in patients with centrifugal continuous-flow left ventricular assist devices. J Heart Lung Transplant. 2016;35:743-751. 26. Lip GYH, Hart RG, Conway DSG. Antithrombotic therapy for atrial fibrillation. BMJ. 2002;325: 1022-1025. 27. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981-992. 28. Suarez J, Patel CB, Felker GM, Becker R, Hernandez AF, Rogers JG. Mechanisms of bleeding and approach to patients with axial-flow
left ventricular assist devices. Circ Heart Fail. 2011; 4:779-784. 29. Meyer AL, Malehsa D, Bara C, et al. Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device. Circ Heart Fail. 2010;3:675-681. 30. Persky RW, Turtzo LC, McCullough LD. Stroke in women: disparities and outcomes. Curr Cardiol Rep. 2010;12:6-13. 31. Poli D, Antonucci E, Grifoni E, Abbate R, Gensini GF, Prisco D. Gender differences in stroke risk of atrial fibrillation patients on oral anticoagulant treatment. Thromb Haemost. 2009;101: 938-942. 32. Paganini-Hill A. Hormone replacement therapy and stroke: risk, protection or no effect? Maturitas. 2001;38:243-261. 33. Harman SM, Brinton EA, Cedars M, et al. KEEPS: The Kronos Early Estrogen Prevention Study. Climacteric. 2005;8:3-12. 34. Feldman D, Pamboukian SV, Teuteberg JJ, et al. The 2013 International Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support: executive summary. J Heart Lung Transplant. 2013;32:157-187. 35. Nassif ME, LaRue SJ, Raymer DS, et al. Relationship between anticoagulation intensity and thrombotic or bleeding outcomes among outpatients with continuous-flow left ventricular assist devices. Circ Heart Fail. 2016;9:e002680.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 25 July 2019; accepted 26 August 2019 Citation: World Neurosurg. (2019) 132:265-272. https://doi.org/10.1016/j.wneu.2019.08.211 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
Outcomes After Intracranial Hemorrhage in Patients with Left Ventricular Assist Devices: A Systematic Review of Literature Theresa Elder1, Alankrita Raghavan4, Arvin Smith4, Christina Huang Wright1,4, James Wright1,4, Christopher Burant5,6, Martha Sajatovic3,2, Alan Hoffer1,4
Key words Hemorrhagic stroke - ICH - Intracranial hemorrhage - LVAD - Ventricular assist device -
Abbreviations and Acronyms ICH: Intracranial hemorrhage INR: International normalized ratio LVAD: Left ventricular assist device From the Departments of 1Neurological Surgery, 2Neurology, and Psychiatry and 3Neurological and Behavioral Outcomes Research Center, University Hospitals Cleveland Medical Center, Cleveland, Ohio; and 4School of Medicine, 5Mandel School of Applied Social Sciences, and 6Frances Payne Bolton School of Nursing, Case Western Reserve University, Cleveland, Ohio, USA To whom correspondence should be addressed: James Wright, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019) 132:265-272. https://doi.org/10.1016/j.wneu.2019.08.211 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
INTRODUCTION Intracranial hemorrhages (ICHs) are common neurosurgical complications associated with high mortality.1-3 Known causes of ICH include hypertensive arteriosclerosis and arteriovenous abnormalities, as well as brain tumors. Patients with left ventricular assist devices (LVADs) are at increased risk of ICH in particular, because of a requirement for continuous anticoagulation to prevent device thrombosis and ischemic stroke.4 The incidence of ICH in patients with LVAD has been reported to fall between 3% and 11% in multiple outcomes studies.5-7 Although LVADs are known to improve survival in patients with advanced heart failure,8 the resultant changes in systemic blood flow patterns cause additional bleeding risks irrespective of anticoagulation requirements, such as
- BACKGROUND:
Intracranial hemorrhage (ICH) is a common complication in patients with left ventricular assist devices (LVADs) and carries a dismal prognosis. Limited data are available on the management and outcomes in this population, which are essential to determine an optimal treatment strategy. We conducted a systematic literature review to determine the clinical characteristics and survival of this population.
- METHODS:
PubMed, Ovid, Embase. Scopus, Cochrane, CENTRAL, and Web of Science articles were selected using the following terms: (“Heart-Assist Devices” or “left ventricular assist device” or “LVAD”) AND (“intracranial hemorrhage” OR “cerebral hemorrhage” OR “brain hemorrhage” OR “intracerebral hemorrhage” OR “intraparenchymal hemorrhage” OR “hemorrhagic stroke”). Abstracts and articles were screened according to inclusion and exclusion criteria that were determined a priori. Potential studies were reviewed by 4 authors, who reached a consensus on the final studies to be included.
- RESULTS:
The literature review yielded 609 abstracts, which were screened according to predetermined inclusion criteria. A total of 143 full-text articles were reviewed, and 8 articles were included in the final qualitative analysis. These studies reviewed data for 597 patients with LVADs who had ICH. The mortality for ICH was widely variable across studies and ranged from 16% to 100%.
- CONCLUSIONS:
There is minimal existing literature on patients with LVAD with ICH that report patient outcomes in a nonstandardized fashion. The studies included in this analysis report mortality consistent with previous reports, indicating a need for further investigation to identify risk factors and improve outcomes in these patients.
acquired von Willebrand disease.9 As a result, patients on LVADs are at increased risk for ICH.10 When LVADs were introduced, they were commonly used as a bridge to transplantation. However, with ongoing donor shortages in combination with advancing technology, an increased proportion of patients are using LVADs as destination therapy.6,11 More than 26,000 patients have received LVAD implants in the last decade, and the number of LVAD implants has shown an average annual increase of 12.6%.12,13 Furthermore, the proportion of patients using LVADs as destination therapy increased from 14.6% in 2007 to 41.6% in 2011, with a recent
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
study citing the average duration of LVAD therapy to be 581 days.14,15 These patients remain on long-term anticoagulation and antiplatelet regimens for longer periods, putting them at further risk for spontaneous bleeding. Previous reports have suggested that as many as 50% of patients on LVADs have bleeding incidents within 30 days of device implantation.16 Gastrointestinal bleeding in patients with LVADs has been well documented in the literature, and with extensive discussion of its management.17 However, ICH risk factors and management strategies have received less attention, with their sole mention in single-institution studies and case series.
www.journals.elsevier.com/world-neurosurgery
265
LITERATURE REVIEW THERESA ELDER ET AL.
Furthermore, these studies often combine patients with ischemic and hemorrhagic stroke, with no distinction in management strategies between the 2 diseases. With as many as two thirds of documented ICH cases resulting in patient death, further investigation is warranted to better understand and treat this complication.4 The goal of our systematic literature review is to understand survival and mortality of ICH in patients with LVADs, specifically investigating the association between anticoagulation regimens and device type with ICH outcome. These data may aid in establishing potential ICH risk assessments for patients with LVADs and help determine optimal treatment strategies for ICH in this patient population. METHODS A systematic review of the reported data was performed in accordance with the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analyses) guidelines. PubMed, Ovid, Embase. Scopus, Cochrane, CENTRAL, and Web of Science articles were selected using the following terms: (“Heart-Assist Devices” or “left ventricular assist device” or “LVAD”) AND (“intracranial hemorrhage” OR “cerebral hemorrhage” OR “brain hemorrhage” OR “intracerebral hemorrhage” OR “intraparenchymal hemorrhage” OR “hemorrhagic stroke”). The reference lists of selected articles were screened for additional articles. Potential studies were reviewed by 4 authors (A.R., A.S., J.W., and C.W.), who reached a consensus on the studies to be included in the present review. The reference lists of all selected reports were screened for additional studies. The article inclusion and exclusion criteria were defined a priori. The inclusion criteria were as follows: original research reports regarding patients with mechanical assist devices who have ICH, regarding data on mortality, morbidity, and discharge deposition, regarding bleeding risk and outcomes with mention of ICH, case series, retrospective and prospective studies, randomized controlled trials, systematic reviews, and clinical trials. The exclusion criteria were as follows: articles regarding only ischemic stroke, case reports on <5 patients, technical notes, letters, review articles, articles in
266
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
languages other than English, animal trials, and in vitro studies. Articles were excluded if they did not address the outcomes of interest (i.e., mortality and survival). Studies were also excluded if the mortality and survival were not clear and/ or they did not explicitly indicate results. All included studies were evaluated by the authors for the quality of evidence provided. Using criteria outlined by Balshem et al.,18 each article was rated into 1 of 4 levels of quality (Table 1). Factors that increased the rating of a study included the size of the study effect, evidence of a dose-response gradient, and lack of confounding variables. Ratings were decreased for inconsistency, indirectness, imprecision, and risk of bias. This rating system was chosen based on its ability to assess an entire body of evidence rather than strictly the evidence of an individual study and helped determine the strength of conclusions drawn. This study was a systematic review and was therefore exempt from institutional review board approval. RESULTS The search yielded 609 abstracts, 143 of which were selected for full-text screening. Several studies were excluded because of failure to provide appropriately calculated mortality or survival metrics. Eight articles were selected for inclusion in the review (Figure 1). All articles published before August 2018 were included in the search. Six studies were retrospective whereas 2 were prospective. All studies were from the United States with the exception of that of Muthiah et al.,25 which was from Australia. Two studies reported external funding: Yavar et al.19 received research funds from Abbot and Medtronic and Muthiah et al.25 received funding from the Pfizer Competitive Cardiovascular Lipid Research Grant and the St. Vincent’s Clinic Grant. The data examined were collected over 14 years (2003e2017), with publication years spanning 2013e2018 (Table 1). These studies reviewed data for of 597 patients with LVADs who had ICH. Acharya et al.20 had the largest sample, describing 423 patients with ICH from Inter-agency Registry for Mechanically Assisted Circulatory Support
(INTERMACS) who received continuousflow LVADs. The mean age of patients studied across all cohorts was 55 years, although 2 of these studies reported the mean age of patients with ICH only. There was a male predominance in the entire population of patients with LVAD as well as in those with ICH, with males comprising 78% of total patients and 64% of the patients with ICH. The incidence of ICH among female patients with LVAD was twice as high as in males, at 12.32% and 6.13%, respectively. Seven studies examined continuous-flow LVADs, whereas one examined both pulsatile and continuous-flow devices. The most frequently implanted device was the Heartmate II (Abbott, Chicago, Illinois, USA), followed by HeartWare (HeartWare International Inc., Framingham, Massachusetts, USA), with no studies including patients with Heartmate III (Abbott) (Table 2). Hemorrhage Characteristics and Analysis of Risk Factors Four studies included the ICH hemorrhage pattern in their data, with intraparenchymal hemorrhage representing most hemorrhages (42%e61%).10,21,22,24 Subarachnoid hemorrhage comprised 24%e48% of all hemorrhages, and both subdural and epidural hemorrhages combined accounted for the remaining 0%e 10% of hemorrhages. Average international normalized ratio (INR) ranged from 1.3 to 3.8 among patients with ICH, with a mean of 2.7 across all studies. All patients were on an anticoagulation regimen at the time of ICH, most of which included warfarin, with the exception of 9 patients,10,22 who were taking either heparin or an alternative nonwarfarin anticoagulant. Six studies reported antiplatelet regimens, all which included aspirin with or without an additional agent, with the exception of 1 patient in Ramey et al.‘s study.10 The mean duration of LVAD therapy before hemorrhage was 454 days. Management and Outcomes Analysis Four studies included the percentage of patients who underwent surgical intervention, which ranged from 0% to 40%.10,21-23 The time points at which mortality and survival were reported varied from same admission mortality to 2-year mortality. Mortality from ICH was widely
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
LITERATURE REVIEW
Low No cohorts, 1 group of patients with LVAD
DISCUSSION
Single Prospective
LVAD, left ventricular assist device.
Australia 2011e2012 Journal of Heart and Lung Transplantation Muthiah et al., 201625
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
varied across studies and ranged from 16% to 100%; however, the 100% mortality was drawn from a population of 2 patients.23 Of the 3 studies reporting 30-day survival, data ranged from 45.3% to 67%, whereas data reported for 2-year survival ranged from 0% to 61% (Table 3).19-21,23
28
Moderate 4-factor prothrombin complex concentrate assisted reversal, reversal with traditional agents, no reversal 2008e2015 ASAIO Wong et al., 201624
United States
Prospective cohort
Single
237
Very low One group of patients with LVAD 2006e2014 ASAIO Chou et al., 201723
United States
Retrospective case series
Single
9
Moderate One group of patients with LVAD with intracranial hemorrhage 2007e2016 World Neurosurgery Tahir et al., 201822
United States
Retrospective cohort
Multicenter
59
Moderate No cohorts, 1 group of patients with LVAD 2003e2012 Journal of Neurosurgery Wilson et al., 201321
United States
Retrospective cohort
Single
330
Low No cohorts, 1 group of patients with LVAD 2012e2015 JACC: Heart Failure Acharya et al., 201720
United States
Retrospective database
Multicenter
7112
Low Any bleeding event, no bleeding event 2008e2017 ASAIO Yavar et al., 201719
United States
Retrospective cohort
Single
375
Moderate Total artificial heart or LVAD 2013e2016 World Neurosurgery Ramey et al., 201710
United States
Retrospective cohort
Single
58
Quality of Evidence Total Participants in Study Number of Centers Study Type Country Years Investigated Journal Title Reference
Table 1. Studies Included in Review
ICH IN PATIENTS WITH LVAD
Cohorts
THERESA ELDER ET AL.
This systematic literature review surveys the available literature on survival and mortality in patients on LVADs with concomitant ICH. As the number of advanced heart failure patients using LVADs continues to increase, more patients will be at risk for ICH as a complication of therapy. Although ICH in the general population is known to carry a poor prognosis, survival data in patients on LVADs are limited and have been reported in a nonstandardized fashion across studies, making risk analysis and outcome determination challenging. ICH is a risk associated with any disease process requiring long-term anticoagulation. However, the risk is disproportionately higher in patients with LVAD compared with other patient populations requiring anticoagulants, such as atrial fibrillation.26 Patients with atrial fibrillation require anticoagulation to target an INR range equal to that of patients with LVAD but have a significantly lower ICH rate, ranging between 0.3% and 0.6% per year.16 A trial by Granger et al.27 compared the effects of various anticoagulants used in atrial fibrillation and cited a 0.47% rate of ICH per year for patients taking warfarin. This trial reported an increase in ICH rate by approximately 40% when aspirin was added to the anticoagulation regimen. Although anticoagulation therapy does play an important role in the increased rate of bleeding complications in the population with LVAD, it is also believed to relate to changes in hemostatic factors causing an acquired von Willebrand disease state.4,28 The nonphysiologic pump hemodynamics place shear stress on circulating blood components, resulting in impaired platelet aggregation similar to that seen in aortic stenosis, but which is reversible on removal of the device.16,29 Continuous-flow LVADs were designed in effort to reduce blood flow pulsatility,
www.journals.elsevier.com/world-neurosurgery
267
LITERATURE REVIEW THERESA ELDER ET AL.
which resulted in an increase in 2-year survival in patients with heart failure using these devices compared with those with pulsatile-flow LVADs.5,7 However, continuous-flow LVADs have been shown to have an increased rate of bleeding complications, with the ICH rate being 8% and 11% for pulsatile-flow and continuous-flow LVADs, respectively.6 Ramey et al.10 noted that 100% of their patient population tested positive for acquired von Willebrand disease, as diagnosed by decreased level of von Willebrand factor multimers, and subsequently emphasized the importance of antiplatelet agent reversal. In another study by Uriel et al.,9 all 31 patients with continuous-flow LVADs were found to have acquired von Willebrand disease, and 58% of these patients had bleeding events. Risk Factors Multiple variables have been investigated as possible risk factors for ICH. Studies stratifying patient outcomes based on hemorrhage pattern consistently reported intraparenchymal hemorrhages to be the most common and also to carry the worst prognosis.10,21,22,24 In Wilson et al.‘s series,21 all 5 ICH cases that necessitated surgical intervention were spontaneous intraparenchymal bleeds. However, the survival at 60 days for these patients was only 20%, bringing into question the survival benefit of operative intervention. Furthermore, most of the reported hemorrhages were spontaneous and carried a worse prognosis compared with traumatic ICH, with the mortality of traumatic subarachnoid hemorrhage reported at 0% in 1 cohort.10,21,22 This trend may be related to the described nonphysiologic hemodynamics caused by the rotary mechanism of the LVADs, leading to platelet dysfunction and increased bleeding risk.28 Another risk factor that emerged in our review was the impact of gender on ICH incidence and mortality. We found that although females composed only 20% of the total patients with LVAD included in this review, they accounted for 36% of the patients with ICH. Yavar et al.19 reported that female patients had a 60% higher risk of hemorrhage compared with male patients, and 5 studies included in this review reflected similar findings, with females having approximately 50%
268
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
increased risk of hemorrhage compared with male patients.19-23 The mechanism behind this characteristic is likely multifactorial. One hypothesis relates to the presence of gender effects on platelet function caused by hormone-related mechanisms.19 There is a known protective effect of endogenous estrogen on stroke rates, because premenopausal women sustain fewer strokes than do age-matched men, whereas postmenopausal women sustain more strokes than do age-matched men.30,31 It is also known that oral contraceptives increase the risk for stroke via hormone-mediated upregulation of endothelial dysfunction.32,33 Considering that most women represented in our review were of postmenopausal age, there is reason to believe that hormone signaling leads to a greater upregulation of LVAD-mediated endothelial dysfunction in female patients than it does in male patients. Although other hypotheses exist that put females at greater risk of bleeding complications
from LVADs, such as decreased body surface area and increased frailty, the exact mechanism remains unclear.33 Two studies in our analysis did not report a difference in hemorrhage risk between genders: Chou et al.23 reported a 1:1 female/male ratio, although their sample size was limited to only 2 patients, whereas Wong et al.24 did not provide a gender breakdown of ICH incidence or mortality. However, even among the studies that did report on gender, there was no significant effect of gender on mortality after ICH. Tahir et al.‘s22 was the largest study to reflect this situation, reporting a total cohort comprised of 32% females, who accounted for 28% of deaths. One commonly reported parameter that was associated with increased mortality was Glasgow Coma Scale score.21 Tahir et al.22 also found that both the extent of midline shift and the size of ICH on initial presentation were associated with poorer outcomes. In that study, patients
Figure 1. PRISMA systematic review flow diagram.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
LITERATURE REVIEW THERESA ELDER ET AL.
ICH IN PATIENTS WITH LVAD
Table 2. Description of Studies
Reference
Mean Age of Entire Study Cohort (years)
Sex (Male/Female) of Entire Study Cohort
Sex (Male/Female) of Patients with ICH Only
Race
Type of LVAD
Ramey et al., 201710
56.7 (patients with ICH only)
49:9
5:0
—
49/58 LVAD (type unspecified), 10/58 TAH, 1/58 both LVAD þ TAH
—
Yavar et al., 201719
55.6 0.7
291:94
24:13
White: 82% Other: 18%
100% continuous
—
Acharya et al., 201720
Not specified
5579:1533
228:135
—
100% continuous
—
Wilson et al., 201321
54 12
253:77
67:33
White: 78%, black: 18%, Asian: 1%, Hispanic: 1%
30% continuous, 70% pulsatile (patients with ICH: 42% continuous, 58% pulsatile)
—
Tahir et al., 201822
54.3 (patients with ICH only)
40:19
40:19
For patients with ICH: white, 54.2%; black, 45.8%
72.2% Heartmate II and continuous, 14.8% Heartware HD and pulsatile, 13% not specified
39 Heartmate II 8 Heartware 7 other
Chou et al., 201723
56.5 12
7:2
1:1
—
100% continuous
6 Heartmate II 3 Heartware
Wong et al., 201624
52 13
—
18:13
—
100% continuous
100% Heartmate II
Muthiah et al., 201625
48 14
25:3
—
—
100% continuous
100% HeartWare
LVAD Brand Name
ICH, intracranial hemorrhage; LVAD, left ventricular assist device; TAH, total artificial heart.
with any amount of midline shift were found to be 35% less likely to survive compared with patients with no shift on initial head computed tomography. Management and Outcomes Anticoagulation and antiplatelet regimens were inconsistently reported across the analyzed studies. In the study by Wilson et al.,21 all patients were taking aspirin and warfarin, which was the suggested anticoagulation regimen in the recent Heartmate II Pivotal Trial, with an INR target range of 2e3. Most studies report a lack of standardized anticoagulation regimen used in their study populations and provide insufficient detail on the regimens used. For example, the study by Uriel et al.9 reported that 68.3% of patients were taking warfarin, 55.7% were taking aspirin, and 58.2% were taking dipyridamole but did not specify which patients received dual-agent therapy, which could significantly affect outcomes. Although long-term anticoagulation regimens varied across the
studies included in this review, those that included INR values in their data did not show a correlation between INR and risk of ICH development. INR goals for patients with LVAD, regardless of the pump type, are generally in the range of 2.0e 3.0.34 Tahir et al.22 found that 37% of their patients with ICH were subtherapeutic with INR <2.0, and Ramey et al.10 found that only 20% of patients with ICH had a supratherapeutic INR. Furthermore, although ICH occurred independently of INR, the anticoagulation reversal strategy used often varied based on INR value at presentation, adding an additional variable that could affect outcome. For example, 66% of patients in the series reported by Tahir et al.22 underwent anticoagulation reversal, whereas in the series reported by Ramey et al.,10 40% of patients received anticoagulation reversal, 40% discontinued anticoagulation, and 20% had no intervention. Use of reversal agents was nonuniform across the included studies. Wilson et al.21
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
consistently reversed antiplatelet agents and warfarin with administration of platelet transfusions and fresh frozen plasma, respectively, whereas Ramey et al.10 used an unspecified combination of fresh frozen plasma, prothrombin complex concentrate, and/or vitamin K with a goal INR <1.4. Tahir et al.22 did not specify their INR reversal strategy but did mention that patients taking novel anticoagulants such as rivaroxoban or dabigatran posed a challenge to rapid reversal. This report was published in 2018, 5 years after Wilson et al.‘s study, highlighting the impact of recent pharmacologic advancements on this clinical scenario. Although most studies recommend reversal or temporary discontinuation of anticoagulation as part of ICH management, the concern for subsequent pump thrombosis and other thrombotic events leads to uncertainty regarding the duration of anticoagulant discontinuation. Even although ICH rate may not correlate with increasing INR, the risk of thrombosis is
www.journals.elsevier.com/world-neurosurgery
269
www.SCIENCEDIRECT.com
Reference
Total Number of Study Incidence Participants of ICH (n/N)
Ramey et al., 201710
58
Hemorrhage Location (%)
Mortality, as Reported (%)
Survival, as Reported (%)
Anticoagulation Regimen (%)
Antiplatelet Regimen (%)
5
Subarachnoid: 40 Intraparenchymal: 60 Epidural/subdural: 0
60
40
80 on warfarin, 20 on heparin
80 on aspirin with or without dipyridamole
Mean Mean Duration International of Left Ventricular Normalized Ratio of % with Assist Device Surgical Before Patients with ICH Intervention Hemorrhage 40
32 months (range, 2e84 months)
2.6
37
—
—
38 at 2 years
100 on warfarin
100 on aspirin
—
158 days
—
7112
423
—
70 at 1 year
45.3 at 1 month 34.8 at 6 months 30.3 at 1 year
—
—
—
—
—
Wilson et al., 201321
330
36
Subarachnoid: 28 33 at 30 days, Intraparenchymal: 47 39 at 1 year Epidural/subdural: 22
45 at 4 years
100 on warfarin
100 on aspirin
14
300 days
—
Tahir et al., 201822
598
59
Subarachnoid: 24 Intraparenchymal: 61 Epidural/subdural: 5
42.4 same admission
—
8.5
470 days
2.85
Chou et al., 201723
9
2
—
100
50 at discharge, 0 at 1 year
1/2 warfarin 1/2 heparin
100 aspirin
0
122 days
1.3
Wong et al., 201624
237
42
Subarachnoid: 48 Intraparenchymal: 42 Epidural/subdural: 10
16
—
—
—
—
578 (445) days
—
Wong et al., 201624 (first cohort)
237
10
—
20
—
—
—
—
661 days
3.8
Wong et al., 201624 (second cohort)
237
10
—
20
—
—
—
—
453 days
3.5
Wong et al., 201624 (third cohort)
237
11
—
9.1
—
—
—
—
618 days
2.4
Muthiah et al., 201625
28
3
—
66.7
—
100 warfarin
100 aspirin and clopidogrel
—
223 days
—
ICH, intracranial hemorrhage.
95 on anticoagulation 95 on unspecified regimen; antiplatelet regimen 88 coumadin, 12 noncoumadin
LITERATURE REVIEW
375
Acharya et al., 201720
ICH IN PATIENTS WITH LVAD
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211
Yavar et al., 201719
THERESA ELDER ET AL.
270 Table 3. Mortality and Intracranial Hemorrhage Data
LITERATURE REVIEW THERESA ELDER ET AL.
inversely related to INR. Nassif et al.35 reported data on 11,000 INRs drawn from patients with LVAD and found that the highest thrombosis rate occurred in patients with INR <1.5, and high rates were also seen with INRs in the 1.5e1.99 range. The rate of thrombotic events was also reported as a function of INR, citing a hazard ratio of 0.39 (95% confidence interval, 0.22e0.70; P ¼ 0.002). When balancing the risk of thrombosis with the risk of bleeding, this study determined the ideal INR goal to be 2.6.35 Ironically, this figure is identical to the average INR seen among the patients with ICH in the study by Ramey et al.,10 reflective of this challenging clinical dilemma. There is no clear understanding of the risk for thrombotic events when aspirin and plavix are withheld or reversed. Wong et al.24 studied the effects of withholding anticoagulation in patients with LVAD with ICH and reported that 11 of their cohort of 31 patients were managed successfully without anticoagulation reversal. However, these patients collectively had a mean ICH volume of 0.4 mL, in contrast to Tahir et al.‘s22 cohort of 19 patients who did not undergo reversal of anticoagulation, who had an average ICH volume of 19 mL. Wilson et al.21 reported no thrombotic events in their study population after reversal of warfarin (with platelets given in 39% of patients and fresh frozen plasma in 61% of patients), with a median duration of suspension for aspirin and warfarin being 6 and 10.5 days, respectively, concluding that aspirin can safely be withheld for 1 week and warfarin can be safely withheld for 10 days. Ramey et al.10 suggested a stepwise algorithm for anticoagulation management, with criteria for reversal based on patients’ neurologic deficit and the size of initial hemorrhage on computed tomography. Per their protocol, only large ICHs (defined as intraparenchymal hemorrhage >3 cm or subarachnoid hemorrhage >1e2 mm thickness) required anticoagulation reversal, but the duration of discontinuation was variable. Furthermore, these recommendations were not applicable to patients in whom surgical intervention was required. The mortality reported in our review was widely variable, partially because of
ICH IN PATIENTS WITH LVAD
inconsistent data collection time points. Some studies defined survival as survival to discharge from the hospital, whereas others reported 1-year and 2-year survival.19-23 Although some studies further stratify survival by functional status, such as the Glasgow Outcome Score or the modified Rankin Score, others do not provide such data.21,22 The lack of standardized reporting challenges the ability to make associations between intervention strategies and outcomes. Future Directions We suggest multiple areas of focus for future studies to help determine an optimal therapeutic approach and improve outcomes. First, patients should be stratified according to the anticoagulation regimen used, specifically distinguishing patients on multiple therapeutic agents from those on single-agent therapy. Second, studies must follow a consistent protocol for anticoagulation reversal, with a specific target INR value as a result of the intervention. Furthermore, the timing of INR reversal should be specified relative to the time of ICH diagnosis, with repeat INR values reported and trended relative to the ICH progression. Outcomes and survival should be measured at consistent time points, ideally at 30 days, 90 days, and 1 year time points, because this will help identify and differentiate between early and late bleeding risks. Limitations The inconsistencies in data reporting contributed to several limitations in our review. Certain variables were reported to affect outcomes in some studies, but were not reported at all in others, such as dual antiplatelet therapy usage or anticoagulation reversal. This factor limited the quality of evidence reported in the studies reviewed as a result. The type of LVAD was often not specified, and Ramey et al.10 included 10 patients with a total artificial heart in their ICH cohort, which affects blood flow hemodynamics and, potentially, the pathophysiologic mechanism of ICH in those patients. Six of the 8 studies were retrospective, and only studies from the United States and Australia provided enough data to be included in our review. Three of the studies contained 5 patients with ICH. These small cohorts contributed to the
WORLD NEUROSURGERY 132: 265-272, DECEMBER 2019
wide ranges reported in outcome measures, particularly the series of 2 patients reported by Chou et al.,23 which showed a 100% 60-day mortality. Although the use of LVADs as destination or longer bridge-to-transplant therapy may primarily be an issue of developed countries, this will become a global issue as more countries undergo the epidemiologic transition. Further data from developing countries are needed to identify regionspecific variables affecting survival. CONCLUSIONS ICH is a common complication in patients with LVAD and carries a grim prognosis. The mortality reported in the studies included in this analysis is consistent with historical reports of mortality in ICH, and further investigation is essential to gaining a better understanding of the prevention and treatment of ICH in patients with LVAD. Future studies should focus on elucidating the association between treatment and survival outcomes, as well as exploring the impact of various anticoagulation regimens on this devastating clinical scenario. REFERENCES 1. Fogelholm R, Murros K, Rissanen A, Avikainen S. Long term survival after primary intracerebral haemorrhage: a retrospective population based study. J Neurol Neurosurg Psychiatry. 2005;76: 1534-1538. 2. Flaherty ML, Haverbusch M, Sekar P, et al. Location and outcome of anticoagulant-associated intracerebral hemorrhage. Neurocrit Care. 2006;5: 197-201. 3. Sacco S, Marini C, Toni D, Olivieri L, Carolei A. Incidence and 10-year survival of intracerebral hemorrhage in a population-based registry. Stroke. 2009;40:394-399. 4. Cho SM, Moazami N, Frontera JA. Stroke and intracranial hemorrhage in HeartMate II and HeartWare left ventricular assist devices: a systematic review. Neurocrit Care. 2017;27:17-25. 5. Miller LW, Pagani FD, Russell SD, et al. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007;357:885-896. 6. Slaughter MS. Long-term continuous flow left ventricular assist device support and end-organ function: prospects for destination therapy. J Card Surg. 2010;25:490-494. 7. Pagani FD, Miller LW, Russell SD, et al. Extended mechanical circulatory support with a continuousflow rotary left ventricular assist device. J Am Coll Cardiol. 2009;54:312-321.
www.journals.elsevier.com/world-neurosurgery
271
LITERATURE REVIEW THERESA ELDER ET AL.
8. Rose EA, Gelijns AC, Moskowitz AJ, et al. Longterm use of a left ventricular assist device for endstage heart failure. N Engl J Med. 2001;345: 1435-1443. 9. Uriel N, Pak SW, Jorde UP, et al. Acquired von Willebrand syndrome after continuous-flow mechanical device support contributes to a high prevalence of bleeding during long-term support and at the time of transplantation. J Am Coll Cardiol. 2010;56:1207-1213. 10. Ramey WL, Basken RL, Walter CM, Khalpey Z, Lemole GM, Dumont TM. Intracranial hemorrhage in patients with durable mechanical circulatory support devices: institutional review and proposed treatment algorithm. World Neurosurg. 2017;108:826-835. 11. Birks EJ. A changing trend toward destination therapy: are we treating the same patients differently? Tex Heart Inst J. 2011;38:552-554. 12. Briasoulis A, Inampudi C, Akintoye E, Adegbala O, Alvarez P, Bhama J. Trends in utilization, mortality, major complications, and cost after left ventricular assist device implantation in the United States (2009 to 2014). Am J Cardiol. 2018; 121:1214-1218. 13. Patel N, Kalra R, Doshi R, Bajaj NS, Arora G, Arora P. Trends and cost of heart transplantation and left ventricular assist devices. JACC: Heart Fail. 2018;6:424. 14. Tsiouris A, Paone G, Nemeh HW, et al. Short and long term outcomes of 200 patients supported by continuous-flow left ventricular assist devices. World J Cardiol. 2015;7:792-800. 15. Health Quality Ontario. Left ventricular assist devices for destination therapy: a health technology assessment. Ont Health Technol Assess Ser. 2016;16: 1-60. 16. Klovaite J, Gustafsson F, Mortensen SA, Sander K, Nielsen LB. Severely impaired von Willebrand factor-dependent platelet aggregation in patients with a continuous-flow left ventricular assist device (HeartMate II). J Am Coll Cardiol. 2009;53: 2162-2167. 17. Cushing K, Kushnir V. Gastrointestinal bleeding following LVAD placement from top to bottom. Dig Dis Sci. 2016;61:1440-1447.
272
www.SCIENCEDIRECT.com
ICH IN PATIENTS WITH LVAD
18. Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64:401-406. 19. Yavar Z, Cowger JA, Moainie SL, Salerno CT, Ravichandran AK. Bleeding complication rates are higher in females after continuous-flow left ventricular assist device implantation. ASAIO J. 2018; 64:748-753. 20. Acharya D, Loyaga-Rendon R, Morgan CJ, et al. INTERMACS analysis of stroke during support with continuous-flow left ventricular assist devices: risk factors and outcomes. JACC Heart Fail. 2017;5:703-711. 21. Wilson TJ, Stetler WR, Al-Holou WN, Sullivan SE, Fletcher JJ. Management of intracranial hemorrhage in patients with left ventricular assist devices. J Neurosurg. 2013;118:1063-1068. 22. Tahir RA, Rotman LE, Davis MC, et al. Intracranial hemorrhage in patients with a left ventricular assist device. World Neurosurg. 2018;113:e714-e721. 23. Chou J, Bermudez C, Kormos R, Teuteberg J. Permanent continuous flow left ventricular assist devices use after acute stabilization for cardiogenic shock in acute myocardial infarction. ASAIO J. 2017;63:e13-e17. 24. Wong JK, Chen PC, Falvey J, et al. Anticoagulation reversal strategies for left ventricular assist device patients presenting with acute intracranial hemorrhage. ASAIO J. 2016;62:552-557. 25. Muthiah K, Connor D, Ly K, et al. Longitudinal changes in hemostatic parameters and reduced pulsatility contribute to non-surgical bleeding in patients with centrifugal continuous-flow left ventricular assist devices. J Heart Lung Transplant. 2016;35:743-751. 26. Lip GYH, Hart RG, Conway DSG. Antithrombotic therapy for atrial fibrillation. BMJ. 2002;325: 1022-1025. 27. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981-992. 28. Suarez J, Patel CB, Felker GM, Becker R, Hernandez AF, Rogers JG. Mechanisms of bleeding and approach to patients with axial-flow
left ventricular assist devices. Circ Heart Fail. 2011; 4:779-784. 29. Meyer AL, Malehsa D, Bara C, et al. Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device. Circ Heart Fail. 2010;3:675-681. 30. Persky RW, Turtzo LC, McCullough LD. Stroke in women: disparities and outcomes. Curr Cardiol Rep. 2010;12:6-13. 31. Poli D, Antonucci E, Grifoni E, Abbate R, Gensini GF, Prisco D. Gender differences in stroke risk of atrial fibrillation patients on oral anticoagulant treatment. Thromb Haemost. 2009;101: 938-942. 32. Paganini-Hill A. Hormone replacement therapy and stroke: risk, protection or no effect? Maturitas. 2001;38:243-261. 33. Harman SM, Brinton EA, Cedars M, et al. KEEPS: The Kronos Early Estrogen Prevention Study. Climacteric. 2005;8:3-12. 34. Feldman D, Pamboukian SV, Teuteberg JJ, et al. The 2013 International Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support: executive summary. J Heart Lung Transplant. 2013;32:157-187. 35. Nassif ME, LaRue SJ, Raymer DS, et al. Relationship between anticoagulation intensity and thrombotic or bleeding outcomes among outpatients with continuous-flow left ventricular assist devices. Circ Heart Fail. 2016;9:e002680.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 25 July 2019; accepted 26 August 2019 Citation: World Neurosurg. (2019) 132:265-272. https://doi.org/10.1016/j.wneu.2019.08.211 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.08.211