Share this article on:

Are Blood Stream Infections Associated With an Increased Risk of Hemorrhagic Stroke in Patients With a Left Ventricular Assist Device?

Aggarwal, Ashim*; Gupta, Ankit*; Kumar, Shivani*; Baumblatt, Jane A.; Pauwaa, Sunil*; Gallagher, Colleen*; Treitman, Adam; Pappas, Pat§; Tatooles, Antone§; Bhat, Geetha*

doi: 10.1097/MAT.0b013e318260c6a6
Clinical Critical Care/Monitoring/Instrumentation

Blood stream infections (BSIs) are an important cause of morbidity and mortality in patients with left ventricular assist devices (LVADs). The aim of this study was to examine the correlation between hemorrhagic cerebrovascular accident (CVA) and BSI after implantation of LVAD for advanced heart failure (HF). This was a retrospective descriptive review of 87 patients with end-stage HF, who underwent implantation of HeartMate II continuous-flow LVAD over a 4 year period. Blood stream infections were diagnosed by serial blood cultures, and suspected neurological complications including CVAs were confirmed by neuroimaging. Extensive patient chart review was performed, and descriptive characteristics were analyzed using SPSS statistical software. The mean age of our study population was 62.3 ± 12.8 years, and the majority of our patients were males (n = 75, 86.2%). The baseline characteristics were comparable in the patients with and without CVAs. Patients with BSI had a much greater incidence of CVA compared to patients without BSI (n = 13, 43.3% vs. n = 5, 10.0%; p < 0.0001). There was an increased mortality in patients with BSI than those without (n = 57, 65.5% vs. n = 30, 34.5%; p = 0.003). The risk of all CVAs (hemorrhagic/ischemic) was eightfold (odds ratio [OR] = 7.9; 95% confidence interval [CI] = 2.4–25.5; p = 0.001] in patients with BSI. Patients with BSI had a >20-fold risk of hemorrhagic CVA (OR = 24; 95% CI = 2.8–201.1; p = 0.03). Advanced HF patients with LVAD support who developed BSI need urgent evaluation and close monitoring for suspected neurological complications, particularly hemorrhagic CVA.

From the *Center for Heart Transplant and Assist Devices, Advocate Christ Medical Center, Oak Lawn, Illinois; †Department of Internal Medicine, University of Illinois in Chicago, Chicago, Illinois; ‡Department of Infectious Diseases, Advocate Christ Medical Center, Oak Lawn, Illinois; and §Department of Cardiothoracic Surgery, Advocate Christ Medical Center, Oak Lawn, Illinois.

Submitted for consideration March 2012; accepted for publication in revised form May 2012.

Disclosures: Antone Tatooles is a speaker and consultant for Thoratec. The other authors have no conflicts of interest to report.

Reprint Requests: Geetha Bhat, PhD, MD, Center for Heart Transplant and Assist Devices, Advocate Christ Medical Center, 4400 W. 95th Street, Oak Lawn, IL 60453. Email:

The use of left ventricular assist devices (LVADs) both as a bridge to transplant (BTT) and destination therapy (DT) improves the quality of life and reduces mortality in patients with advanced heart failure (HF).1 Left ventricular assist device therapy has evolved into a standard therapy for patients with advanced HF, who do not get transplanted.

Mechanical circulatory support (MCS) can be associated with significant risk of complications such as thromboembolism,2 infection,3 device failure,2 bleeding,2 and cerebrovascular accidents (CVAs).4 The development of these complications could potentially delay or exclude patients from the transplant list and add major mortality and morbidity. Cerebrovascular accidents (ischemic/hemorrhagic) in any population can have devastating consequences. An increased incidence of ischemic (ICVA) and hemorrhagic CVA (HCVA) after LVAD placement of 8–25% has been reported.5–7 In the LVAD population, the CVAs account for approximately 9% of mortality. In our clinical experience, there was an increased incidence of spontaneous intracranial bleeds despite therapeutic anticoagulation especially in patients with preceding blood stream infections (BSIs). Studies have demonstrated an increased risk of cerebral infarction in bacteremic patients compared to the general population of the same age. There, however, is no literature of a similar association in patients with LVADs.

The aim of our study was to determine the incidence of CVAs (ICVA/HCVA) in our patient population of continuous-flow LVADs (CF-LVADs) and to evaluate a possible association with infectious etiology.

Back to Top | Article Outline


Study Population and Description

This study consisted of a retrospective chart review of 90 consecutive patients who were implanted with the HeartMate II (HMII) between 2005 and 2009. Three patients with transient bacteremia were excluded. All the patients had end-stage HF, requiring implantation of the HMII as a BTT or DT.

Back to Top | Article Outline

Data Collection

Baseline demographic, clinical, and microbiological data were gathered from electronic medical records. Demographic characteristics included age, gender, body mass index, and race. Patients’ medical histories were reviewed for comorbidities including hypertension, diabetes mellitus, atrial fibrillation (AF), and a previous history of CVA. Laboratory data including platelet counts and international normalized ratio (INR) were collected and analyzed. Cerebrovascular accidents were categorized as either HCVA or ICVA and had to be confirmed by computed tomography scan. Blood stream infections were determined by laboratory confirmation per the Centers for Disease Control and National Healthcare Safety Network definition.8 Device infection was defined as a culture-positive specimen obtained from any part of the device, including driveline or pocket infections. The organism causing device infection had to be the same as the one isolated from the BSI, to be included in the analysis. Patients were excluded if they had transient bacteremia with subsequent negative blood cultures before treatment or if an infection did not result in BSI. Patients who developed a CVA before having a BSI or >6 months after BSI were excluded.

Patients were treated with appropriate antibiotics at the time of their documented infection. The duration of device support was calculated from the implantation date to the date of data collection or death. Mortality data were also collected. The study was approved by the Institutional Review Board of our hospital.

Back to Top | Article Outline

Statistical Analysis

Data analysis included both descriptive statistics of the study population and inferential statistical analysis. Categorical variables were reported as number and percentage and analyzed using χ2 test. Continuous variables were reported as mean ± SD, and one-way analysis of variance with post hoc Tukey or Bonferroni was used to compare the three groups: HCVA, ICVA, and no CVA. A forward stepwise logistic ­regression with univariate screening was performed to determine whether the selected variables were significant predictors of the HCVA/ICVA in two separate regression models. p < 0.05 was considered significant. Confidence intervals (CIs) are reported as 95% confidence limits. Kaplan-Meier survival analysis was done to compare the three groups. All analysis was done using SPSS V 11.5 statistical software (SPSS Inc., Chicago, IL).

Back to Top | Article Outline


There were 87 patients included in the study. The mean age of the cohort was 62.3 ± 12.8 years (27–83 years). Majority of the patients were males (n = 75, 86.2%). The mean duration of device support was 576.1 ± 430.1 days (30–1670 days). There were a total of 30 (34.4%) patients with documented BSI after implantation of LVAD. The causes of deaths included sepsis with multiorgan failure in 15 (33.3%), HCVA in 10 (22.2%), device malfunction in 5 (11.1%), ICVA in 4 (8.8%), and gastrointestinal bleed in 3 (6.6%) patients, whereas 8 (17.7%) patients had unknown causes of out-of-hospital deaths.

Table 1 lists the baseline characteristics of the patient population based on CVA (HCVA/ICVA/no CVA). There was no statistical difference in the demographics or comorbidities (AF and history of CVA) between the groups. There was no difference in the INR or the platelet counts between the groups. Temporary device support with the intra-aortic balloon pump or the extracorporeal membrane oxygenator did not differ in the groups. None of the patients required support with right ventricular assist device. There was a statistically significant difference noted in the incidence of BSI and all-cause mortality between the patient groups.

Table 2 shows the baseline characteristics based on the presence or absence of BSI. There was no difference in the demographics between the two groups. There was greater mortality in patients with BSIs than those without BSIs (n = 57, 65.5% vs. n = 30, 34.5%; p = 0.003). A CVA rate of 43.3% (n = 13) was found in patients with BSIs and 10.0% (n = 5) in patients without BSIs (p < 0.0001).

Microbiological data are listed in Table 3. In the 30 cases of BSI, 34 organisms (bacteria/fungi) were identified. Twenty-six of the 30 (86%) patients had BSI caused by a single organism and 4 (14%) patients had infections caused by two organisms. The most commonly identified organism was coagulase-negative Staphylococcus aureus that was found in 16 (47.1%) cases. In the subgroup of patients with infections by two causative organisms, coagulase-negative Staphylococcus aureus was isolated in all patients with the second organism being one of the following: methicillin-sensitive Staphylococcus aureus, Pseudomonas aeruginosa, Achromobacter xylosoxidans, and Corynebacterium diphtheria. Candida species and Erythrocera facialis were the next most likely causes of infections, occurring in 8.8% of infections. To treat these BSI, antibiotics were prescribed as per the recommendations of Infectious Disease specialty.

In our cohort, a total of 10 (11.4%) patients had HCVA. Nine of the 10 (90.0%) patients had HCVA within 6 months of BSI. The average amount of time between the BSI and CVA was 51.4 ± 66.7 days (1–172 days). In HCVA group, the median interval between BSI and CVA was 12 days (25th to75th percentile: 1–117.5 days), whereas in ICVA group the median interval between BSI and CVA was 49 days (25th to 75th percentile: 2.5–107.5 days). Five of the 10 (50%) patients with HCVA underwent urgent neurosurgery. All patients (100%) with HCVA, regardless of BSI, had a poor outcome and died. The mean survival time after CVA was 2.6 ± 1.8 days (1–6 days). The mean duration between LVAD implantation and HCVA was 624.1 ± 392.0 days (113–1222 days). Eight (10%) patients had ICVA and four (50%) had the ICVA within 6 months of BSI. The mean time between the BSI and ICVA was 40.0 ± 63.3 days (1–113 days). Six of these eight (75%) patients with ICVA died. Death after ICVA occurred after 44.1 ± 59.3 days (1–131 days). The two patients who survived have lived for over 560 days after ICVA. Kaplan-Meier survival curve (Figure 1) demonstrates survival curves among patients with HCVA, ICVA, and no CVA. There was a significant survival advantage in patients with no CVA compared to patients with HCVA (p < 0.0001) or ICVA (p = 0.001). However, no difference in survival rate was detected in patients with ICVA and HCVA (p = 0.83).

All patients were on anticoagulation and antiplatelet regimen, which consisted of warfarin (target INR of 1.5–2.5) and aspirin 325 mg. The average INR for the patients who had a HCVA was 1.8 ± 0.6 (1.1–2.6).

Results of multivariate regression models showed that the risk of all CVAs (HCVA/ICVA) was eightfold (odds ratio [OR] = 7.9; 95% CI = 2.4–25.5; p = 0.001] in patient with BSI compared to ones with no BSI. In addition, patients with BSI were approximately 24 times more likely to suffer HCVA (OR = 24; 95% CI = 2.8–201.1; p = 0.03). There was no significant increase in the risk of ICVA in patients with BSIs when analyzed in a separate regression model. None of the other variables (previous history of CVA and AF) studied was associated with a statistically significant increase in the risk of CVA in patients with BSI.

Back to Top | Article Outline


The results of our study suggest that LVAD patients with post-implant BSIs are at an increased risk of CVAs. The incidence of CVA of 10% found in our patient population without BSIs is comparable to 9% as reported by the The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) registry data9; however, this risk quadrupled (44%) in our patients with BSIs. This difference was statistically significant (p < 0.0001). An increased incidence of CVA in bacteremia has been studied previously.10–12 Evidence from the endocarditis literature suggests that 20–30% of patients with endocarditis may suffer CVAs.13 In addition, studies have demonstrated an increased risk of cerebral infarction in bacteremic patients compared to the general population of the same age.12,14

The median time to HCVA after BSI was 12 days compared to 49 days for ICVA in our study population. As most CVAs occurred within 60 days after BSI in our study, patients with neurological signs or symptoms need urgent evaluation and close monitoring for adverse neurological events. In our population of LVAD recipients, this risk of combined end-point of ICVA/HCVA was about eight times higher in patients with BSIs than patients without BSIs and patients with BSIs had a higher risk for a HCVA (OR = 24). Blood stream infections, however, did not seem to be an independent predictor of ICVA. To the best of our knowledge, this is the first study demonstrating an increased risk of HCVA in LVAD patients with BSIs.

Several studies have demonstrated stimulation of the inflammatory system after implantation of various types of LVADs.15,16 Although the survival rate is significantly improved with long-term LVAD support, the immunologic response is significantly accelerated in these patients, thereby predisposing them to opportunistic infections.3 The various pathophysiologic mechanisms such as the activation of leukocyte counts, complement system, and acute phase proteins have been implicated in the etiology of microemboli and ICVA.17,18 Kato et al.19 in their recent study of 307 patients post-LVAD surgery (167 HeartMate I and 140 HMII devices) demonstrated LVAD infection to be an independent risk factor for CVA in addition to other noninfectious factors.

Studies have shown that the percentage of patients who experience an intracranial hemorrhage (ICH) while on LVAD is approximately 13–14%,20,21 making it a well-known and feared risk of MCS. Anticoagulation and antiplatelet therapy is required in patients after implantation of LVAD, which can increase the risk of bleeding. It is also thought that this risk is augmented by the effects of being on cardiopulmonary bypass, which causes platelet dysfunction, platelet consumption, and hemodilution of clotting factors.22 Specifically, it has also been shown that female sex, creatinine >2.6 mg/dl, and the need for dialysis are independent risk factors for ICH.21 For LVADs, the specific risk factors include female sex, older age, and required biventricular support.20 However, none of these factors except BSI seemed to affect the incidence of HCVA in our study. The rate of mortality from MCS dramatically increases from 61% to 92.3% when ICH develops,23,24 as has been demonstrated in our study (100%).

The pathogenesis of HCVA in LVAD patients is not completely understood and may be hypothesized as follows: 1) Migration of embolic fragments from the original site of obstruction in the brain microvasculature allows reperfusion of the previously infarcted tissue.25 If capillaries in these reperfused areas are severely damaged, hemorrhagic transformation can occur.26 In anticoagulated patients, hemorrhagic transformation more often results in disastrous neurological worsening. 2) The second hypothesis is the formation of cerebral mycotic aneurysms from bacterial seeding during BSI, which may eventually rupture and bleed. 3) Weakening of the vessel walls in the brain microvasculature as a result of infection and inflammation (infectious vasculitis), which might eventually bleed.

The emergent management in patients with HCVA obviates the need for immediate reversal of anticoagulation. It is essential to immediately reverse the patient’s anticoagulation and antiplatelet therapy with fresh frozen plasma, cryoprecipitate, factor VII, platelets, protamine, or vitamin K (specific to the therapy the patient was receiving)27–29 to minimize the risk of hemorrhagic expansion, especially in the setting of life-threatening ICH. Although no standard protocol for restarting anticoagulation or antiplatelet therapy after a neurosurgical procedure in this patient population is found in the literature, guidelines generally state that the decision should be patient specific and individualized by the treating medical team and wishes of the patient and family.27,28 In patients who are at high risk for thromboembolism, in whom reinitiating anticoagulation is seemingly necessary, anticoagulation or antiplatelet therapy may be started 7–14 days after ICH has stopped.28,30,31 The results of this study suggest that BSIs are an independent predictor of CVA in general and more importantly HCVA in LVAD patients. Because this is a pilot study, it raises more questions than it answers. Thus, a larger, more comprehensive study will need to be designed, with a larger population of patients. Given the small sample size, a weakness of this study was that it did not take into account a comprehensive list of factors that would predispose these patients to CVAs such as history of smoking, diabetes mellitus, and blood pressures before LVAD implantation and before CVA. In addition, it is worth noting that the clinical course of each patient after LVAD implantation may have been markedly different, and there may have been factors associated with increased CVA risk other than the ones available for this study that may have been significant. The results from this study are limited due to its retrospective design and being from a single center.

Back to Top | Article Outline


In conclusion, this study suggests that, in patients with CF-LVADs, BSI increases the risk of CVA and more importantly is an independent predictor of HCVA. Patients with BSI should be closely monitored for any neurological signs/symptoms, and early neuroimaging may help prompt the detection of catastrophic life-threatening neurologic events.

Back to Top | Article Outline


1. Miller LW, Pagani FD, Russell SD, et al.HeartMate II Clinical Investigators. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007;357:885–896
2. Slaughter MS, Pagani FD, Rogers JG, et al. Clinical management of continuous-flow left ventricular assist devices in advanced heart failure. J Heart Lung Transplant. 2010;29:S1–S39
3. Holman WL, Park SJ, Long JW, et al.REMATCH Investigators. Infection in permanent circulatory support: Experience from the REMATCH trial. J Heart Lung Transplant. 2004;23:1359–1365
4. Pae WE, Connell JM, Boehmer JP, et al. Neurologic events with a totally implantable left ventricular assist device: European LionHeart Clinical Utility Baseline Study (CUBS). J Heart Lung Transplant. 2007;26:1–8
5. Slaughter MS, Rogers JG, Milano CA, et al.HeartMate II Investigators. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009;361:2241–2251
6. John R, Kamdar F, Liao K, Colvin-Adams M, Boyle A, Joyce L. Improved survival and decreasing incidence of adverse events with the HeartMate II left ventricular assist device as bridge-to-transplant therapy. Ann Thorac Surg. 2008;86:1227–1234; discussion 1234
7. Tsukui H, Abla A, Teuteberg JJ, et al. Cerebrovascular accidents in patients with a ventricular assist device. J Thorac Cardiovasc Surg. 2007;134:114–123
8. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control. 2008;36:309–332
9. Kirklin JK, Naftel DC, Kormos RL, et al. Third INTERMACS Annual Report: The evolution of destination therapy in the United States. J Heart Lung Transplant. 2011;30:115–123
10. Valtonen V, Kuikka A, Syrjänen J. Thrombo-embolic complications in bacteraemic infections. Eur Heart J. 1993;14(suppl K):20–23
11. Syrjänen J, Valtonen VV, Iivanainen M, Kaste M, Huttunen JK. Preceding infection as an important risk factor for ischaemic brain infarction in young and middle aged patients. Br Med J (Clin Res Ed). 1988;296:1156–1160
12. Syrjänen J, Valtonen VV, Iivanainen M, Hovi T, Malkamäki M, Mäkelä PH. Association between cerebral infarction and increased serum bacterial antibody levels in young adults. Acta Neurol Scand. 1986;73:273–278
13. Matsushita K, Kuriyama Y, Sawada T, et al. Hemorrhagic and ischemic cerebrovascular complications of active infective endocarditis of native valve. Eur Neurol. 1993;33:267–274
14. Lindsberg PJ, Grau AJ. Inflammation and infections as risk factors for ischemic stroke. Stroke. 2003;34:2518–2532
15. Loebe M, Koster A, Sänger S, et al. Inflammatory response after implantation of a left ventricular assist device: Comparison between the axial flow MicroMed DeBakey VAD and the pulsatile Novacor device. ASAIO J. 2001;47:272–274
16. Ankersmit HJ, Wieselthaler G, Moser B, et al. Transitory immunologic response after implantation of the DeBakey VAD continuous-axial-flow pump. J Thorac Cardiovasc Surg. 2002;123:557–561
17. Gu YJ, Mariani MA, Boonstra PW, Grandjean JG, van Oeveren W. Complement activation in coronary artery bypass grafting patients without cardiopulmonary bypass: The role of tissue injury by surgical incision. Chest. 1999;116:892–898
18. Edmunds LH Jr. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg. 1998;66(5 suppl):S12–S16; discussion S25
19. Kato TS, Schulze PC, Yang J, et al. Pre-operative and post-operative risk factors associated with neurologic complications in patients with advanced heart failure supported by a left ventricular assist device. J Heart Lung Transplant. 2012;31:1–8
20. Deng MC, Edwards LB, Hertz MI, et al.International Society for Heart and Lung Transplantation. Mechanical circulatory support device database of the International Society for Heart and Lung Transplantation: Third annual report–2005. J Heart Lung Transplant. 2005;24:1182–1187
21. Kartha V, Gomez W, Wu B, Tremper K. Laparoscopic cholecystectomy in a patient with an implantable left ventricular assist device. Br J Anaesth. 2008;100:652–655
22. Hampton CR, Verrier ED. Systemic consequences of ventricular assist devices: Alterations of coagulation, immune function, inflammation, and the neuroendocrine system. Artif Organs. 2002;26:902–908
23. Kasirajan V, Smedira NG, McCarthy JF, Casselman F, Boparai N, McCarthy PM. Risk factors for intracranial hemorrhage in adults on extracorporeal membrane oxygenation. Eur J Cardiothorac Surg. 1999;15:508–514
24. Gruber EM, Seitelberger R, Mares P, Hiesmayr MJ. Ventricular thrombus and subarachnoid bleeding during support with ventricular assist devices. Ann Thorac Surg. 1999;67:1778–1780
25. Hart RG, Sherman DG. Stroke and the total artificial heart: Neurologic considerations. Tex Heart Inst J. 1987;14:63–71
26. Hart RG, Easton JD. Hemorrhagic infarcts. Stroke. 1986;17:586–589
27. Testai FD, Aiyagari V. Acute hemorrhagic stroke pathophysiology and medical interventions: Blood pressure control, management of anticoagulant-associated brain hemorrhage and general management principles. Neurol Clin. 2008;26:963–985, viii
28. Broderick J, Connolly S, Feldmann E, et al.American Heart Association/American Stroke Association Stroke Council. American Heart Association/American Stroke Association High Blood Pressure Research Council. Quality of Care and Outcomes in Research Interdisciplinary Working Group. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: A guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Circulation. 2007;116:e391–e413
29. Thunberg CA, Gaitan BD, Arabia FA, Cole DJ, Grigore AM. Ventricular assist devices today and tomorrow. J Cardiothorac Vasc Anesth. 2010;24:656–680
30. Aguilar MI, Hart RG, Kase CS, et al. Treatment of warfarin-associated intracerebral hemorrhage: Literature review and expert opinion. Mayo Clin Proc. 2007;82:82–92
31. Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage associated with oral anticoagulant therapy: Current practices and unresolved questions. Stroke. 2006;37:256–262
Copyright © 2012 by the American Society for Artificial Internal Organs