Activation-induced T-cell death and immune dysfunction after implantation of left-ventricular assist de vice

Activation-induced T-cell death and immune dysfunction after implantation of left-ventricular assist de vice Hendrik Jan Ankersmit, Sorina Tugulea, Talia Spanier, Alan D Weinberg, John H Artrip, Elizabeth M Burke, Margaret Flannery, Donna Mancini, Eric A Rose, Niloo M Edwards, Mehmet C Oz, Silviu Itescu

Summary Background Cardiac transplantation is a limited option for end-stage heart failure because of the shortage of donor organs. Left-ventricular assist devices (LVADs) are currently under investigation as permanent therapy for end-stage heart failure, but long-term successful device implantation is limited because of a high rate of serious infections. To examine the relation between LVAD-related infection and host immunity, we investigated immune responses in LVAD recipients. Methods We compared the rate of candidal infection in 78 patients with New York Heart Association class IV heart failure who received either an LVAD (n=40) or medical management (controls, n=38). Fluorochrome-labelled monoclonal antibodies were used in analyses of T-cell phenotype. Analysis of T-cell function included intradermal responses to recall antigens and proliferative responses after stimulation by phytohaemagglutinin, monoclonal antibodies to CD3, and mixed lymphocyte culture. We measured T-cell apoptosis in vivo by annexin V binding, and confirmed the result by assessment of DNA fragmentation. Activation-induced T-cell death was measured after T-cell stimulation with antibodies to CD3. All immunological tests were done at least 1 month after LVAD implantation. Between-group comparisons were by Kaplan-Meier actuarial analysis and Student’s t test. Findings By 3 months after implantation of LVAD, the risk of developing candidal infection was 28% in LVAD recipients, compared with 3% in controls (p=0·003). LVAD recipients had cutaneous anergy to recall antigens and lower (<70%) T-cell proliferative responses than controls after activation via the T-cell receptor complex (p<0·001). T cells from LVAD recipients had higher surface expression of CD95 (Fas) (p<0·001) and a higher rate of spontaneous apoptosis (p<0·001) than controls. Moreover, after stimulation with antibodies to CD3, CD4 T-cell death increased by 3·2-fold in LVAD recipients compared with only 1·2-fold in controls (p<0·05). Interpretation LVAD implantation results in an aberrant state of T-cell activation, heightened susceptibility of CD4 T cells to activation-induced cell death, progressive defects in cellular immunity, and increased risk of opportunistic infection. Lancet 1999; 354: 550–55 Departments of Surgery (H J Ankersmit MD , T Spanier MD , A D Weinberg MS, J H Artrip MD, E M Burke RN, M Flannery RN , Prof E A Rose MD, N M Edwards MD , M C Oz MD, S Itescu MD ), Medicine (D Mancini MD), and Pathology (S Tugulea MD), College of Physicians and Surgeons of Columbia University, New York, NY 10032, USA Correspondence to: Dr Silviu Itescu (e-mail: [email protected])

Introduction Congestive heart failure is a major public-health issue.1 Each year more than 60 000 patients in the USA are candidates for cardiac transplantation owing to 2-year mortality rates of 70–80%.2 However, the limited supply of donor organs means that fewer than 3000 cardiac transplants are performed annually.3,4 Left-ventricular assist devices (LVAD) are increasingly used as bridges to cardiac transplantation, with encouraging mediumterm results.5–7 Prospective, randomised, multicentre trials of permanent LVAD implantation for patients with end-stage heart failure are underway. The clinical success of LVAD implantation has, nevertheless, been accompanied by complications such as thromboembolic events in as many as 30% of cases.8,9 This complication is reduced to less than 4%10 in types of LVAD that incorporate a textured surface lining which supports the growth of neointima-type cells.11,12 The most serious complication, irrespective of the type of LVAD used, is the high rate of systemic infections, which has been reported in 25–48% of recipients.13,14 Although some of the infections can be attributed to drive line, pocket, and perioperative complictions, the frequent identification of bacteraemia and, particularly, of fungal infection in LVAD recipients15–19 raises the possibility that LVAD implantation may be associated with the development of defects in host immunity, as reported after total artificial heart implantation.20 Indeed, the risk of fungal infection has been deemed so high that prophylactic use of antifungal therapy in selected LVAD recipients has been advocated.15 We aimed to compare the rates of candidal infection in LVAD recipients and in a control group of patients with heart failure who were treated medically, and to examine cellular immune function in these groups.

Methods Candidal infection We examined the rate of candidal infection in 78 patients with New York Heart Association (NYHA) class IV heart failure who were awaiting cardiac transplantation. All the patients were classified as United Network of Organ Sharing (UNOS) status I. 40 patients received a Thermo Cardiosystems Inc (TCI) Heartmate (TCI, Boston, MA, USA) left-ventricular assist device (LVAD) between 1992–1996, and 38 patients had medical management (the control group). Candidal infection was defined as the presence of positive cultures, obtained when medically warranted, from blood or extravascular sites. Surveillance cultures were not done routinely. Among the LVAD recipients, the median duration of implantation was 138 (range 21–412) days, their median age was 51 (18–65) years, 33 were men, and 35 had ischaemic cardiomyopathy. Among the controls, the median age was 52 (20–64) years, 33 were men, and 34 had ischaemic cardiomyopathy. The institutional review board of our hospital approved the study and we obtained the informed consent of all the patients who took part in the study.

T-cell apoptosis in vivo Flow cytometry—PBMCs were analysed with a Flow Cytometric Apoptosis Detection Kit (Becton Dickinson Systems, McKinley, MN, USA). 3⫻105 PBMC were stained with phycoerythrin-conjugated monoclonal antibodies to CD3, CD4, or CD8, and then costained with 10 µL fluoresceinisothiocyanate-conjugated annexin V (R&D systems, Minneapolis, MN, USA) to detect phospatidylserine expression on cells during early apoptotic phases.22,23 The samples were analysed by FACStar 500. Analysis of DNA fragmentation by 3’ end labelling of DNA— DNA fragmentation was measured by radiolabelling technique, as previously described.24 Briefly, DNA was obtained by standard techniques from 5⫻106 PBMC of six LVAD recipients and three controls. Radiolabelled dideoxynucleotide ([␣ 32P]-ddATP: 11·1⫻107 MBq/mmoL) was added to each DNA sample by terminal deoxynucleotidyl transferase, under conditions that ensured only one molecule of ddATP was incorporated per 3’ end. The labelled samples were loaded into 2% agarose gels, separated by electrophoresis, and exposed for radiography for 6 h at ⫺70°.

Activation-induced cell death after stimulation of T-cell receptor in vitro

Figure 1: Kaplan-Meier time-dependent analysis of development of candidal infection in recipients of LVADs and controls

Cellular immune function Immunological investigations were done in 14 LVAD recipients, 12 of whom had a TCI device and two had a Novacor device (Baxter Health Care Corp, IL, USA), and in 20 controls. No patients had any infection during the study and none received immunosuppression. All immunological tests were done at least 1 month after implantation to limit the confounding effects of surgery.

Immunophenotype of circulating T cells Cell counts from LVAD recipients and controls were done by Coulter Counter analysis (Fullerton, CA, USA). We used fluorochrome-labelled monoclonal antibodies to CD4, CD8, and CD95 in two-colour or three-colour immunoflourescence analyses, as previously described by Robins.21 Log fluorescence was measured by a FACScan 500 flow cytometer (Becton Dickinson Systems, San Jose, CA, USA).

In-vitro and in-vivo T-cell function For in-vitro studies of T-cell function, we used Ficoll to isolate peripheral-blood mononuclear cells (PBMC) from samples of whole blood that had been anticoagulated with edetic acid. 5⫻105 PBMC were added to 96-well flat-bottomed plates in 200 µL Roswell Park Memorial Institute medium. For blastogenesis assays, cells were stimulated with phytohaemagglutinin-P (4 µg/mL) or soluble monoclonal antibodies to CD3 (10 ng/well) for 48 h at 37°C. For mixedlymphocyte-culture assays, responder cells (5⫻105 cells) were added to irradiated (25 Gy) stimulator cells (5⫻105 cells) or medium alone and cultured for 5 days at 37°C. Cells were pulsed for 16 h with [3H] thymidine (3·7⫻104 Bq/well) after 48 h in the blastogenesis assays and after 5 days in the mixed lymphocyte culture. Cultures were harvested and measured in a liquid scintillation counter. For in-vivo studies of T-cell function, 0·1 mL samples of mumps skin test antigen and Candida albicans skin test antigen were injected intradermally in LVAD recipients (>1 month after implantation) or in controls. 48 h after the injection, the induration responses were read by an investigator. An induration response of greater than 5 mm was deemed positive.

1⫻106 PBMC from five LVAD recipients and four controls were placed in 48-well plates and cultured with monoclonal antibodies to CD3 (50 ng/mL) or isotype control antibody for 24 h at 37°C. Resting or stimulated T cells were stained simultaneously with fluorochrome-conjugated monoclonal antibodies to CD4 or CD8, annexin V, and propidium iodide. We used cytometry by FACStar 500 to detect apoptotic cells that were undergoing cell death after T-cell-receptor engagement.

Statistical analyses Between-group differences in the rate of disseminated candidal infections were compared by Kaplan-Meier actuarial analysis, with p values calculated by log-rank statistics. We defined time zero (baseline) for both groups as the date at which patients were listed as UNOS status I awaiting cardiac transplantation; in all LVAD recipients, time zero coincided with the date of implantation. Continuous variables, such as proliferative responses to various stimuli, were analysed by Student’s t test.

Results Rate of disseminated candidal infections LVAD implantation was associated with a significantly increased risk of developing candidal infection. Figure 1 shows that the risk of developing candidal infection within the first 3 months after LVAD implantation was 28% for LVAD recipients, compared with only 3% in

Figure 2: Spontaneous proliferative activity of circulating T cells in LVAD recipients and controls Results are mean of stimulation indices from six LVAD recipients at less than 50 days since implantation, six LVAD recipients at more than 50 days since implantation, and of 15 controls. Bars are SE.

Figure 3: T-cell proliferative responses to stimulation by mixed lymphocyte culture, monoclonal antibodies to CD3, and phytohaemagglutinin in LVAD recipients and controls Results are mean of stimulation indices from 12 LVAD recipients and 15 controls. Bars are SE.

the controls (p<0·01). This rate was also significantly higher than the rate of candidal infection in other patients undergoing cardiac surgery (<1%).25 We then investigated whether the observed increase in fungal infection in LVAD recipients was related to reduced in-vivo cellular responsiveness to recall antigens. After intradermal challenge with mumps antigens, one of seven LVAD recipients showed reactivity compared with all ten controls who were challenged (p<0·001). Moreover, none of seven LVAD recipients reacted to intradermal challenge with candidal antigens, compared with five of ten controls (p<0·05). These findings confirm the presence of in-vivo T-cell defects in LVAD recipients.

Figure 4: T-cell expression of (A) CD95(Fas) and (B) annexin V binding, which indicates spontaneous apoptosis, in LVAD recipients and controls Mean of values from 12 LVAD recipients and 20 controls. Bars are SE.

Proliferative activity of T cells Figure 2 shows that spontaneous proliferactive activity was significantly greater in the T cells of LVAD recipients than in those of the controls, and that this activity increased with duration of LVAD implantation. By contrast with this spontaneous proliferative activity, T cells from LVAD recipients showed defective proliferative responses after activation, specifically via the T-cell receptor complex (figure 3). After allogeneic mixed lymphocyte culture, the mean stimulation index of T cells from LVAD recipients was 74% lower than that of T cells from the controls (1·9 [SD 0·2] vs 7·4 [1·1], p<0·001; figure 3). Similarly, after activation with monoclonal antibodies to CD3, the mean stimulation index of T cells from LVAD recipients was 83% lower than that of T cells from the controls (9·7 [5·8] vs 57·1 [19·3]; p<0·001; figure 3). By contrast, activation of T

Figure 5: Fragmentation of DNA from T cells of two LVAD recipients (lane 3, 4) and one control (lane 2) Lane 1 is size-marker. Lane 3 at day 60 after implantation. Lane 4 at day 98 after implantation. Fragmentation is greater in LVAD recipients than in control.

Figure 7: Proportion of annexin-V-binding CD4 T cells that underwent cell death after stimulation by monoclonal antibodies to CD3 in LVAD recipients and controls Figure 6: Flow cytometry of proportion of CD4 T cells in representative LVAD recipient and control that underwent apoptosis (annexin V binding) and cell death (propidum iodide) after 24 h of culture with medium or monoclonal antibodies to CD3

cells by pathways other than stimulation of T-cell receptor caused similar increases in T-cell proliferation in both groups. Figure 3 shows that after activation by phytohaemagglutinin, the mean stimulation index for LVAD recipients was 44·9 (11·3) compared with 50·8 (12·1) for controls (p>0·5). Expression of CD95 (Fas) and spontaneous T-cell apoptosis in vivo To assess whether the increased spontaneous proliferative activity of circulating T cells in LVAD recipients was a result of increased T-cell activation in vivo, we measured both T-cell activation and surface expression of CD95, a molecule associated with a pathway of cellular apoptosis, in LVAD recipients and controls. Figure 4 shows that T cells from LVAD recipients expressed significantly higher surface concentrations of CD95 (Fas) than T cells from the controls. Mean expression of CD95 was increased to a similar degree in both CD4 and CD8 T cells from the LVAD recipients compared with the controls (70 [6] vs 22 [4]%, p<0·001 and 69 [7] vs 7 [2]%, p<0·001, respectively). We then examined whether this heightened state of T-cell activation in LVAD recipients was associated with increased T-cell apoptosis in vivo. CD4 and CD8 T cells from LVAD recipients had higher levels of annex in V binding than CD4 and CD8 T cells from the controls (39 [5] vs 4 [1]%, p<0·001 and 45 [4] vs 2 [1]%, p<0·001, respectively; figure 4). This increased state of T-cell apoptosis in vivo was confirmed by analysis of DNA isolated from freshly obtained T cells. As shown in figure 5, a characteristic fragmentation pattern of apoptosis was observed in DNA from circulating T cells of the LVAD recipients, but not in DNA from T cells of any controls. No differences in T-cell CD95 expression or apoptosis were found between patients with TCI or Novacor devices.

Results are mean of values obtained from five LVAD recipients and four controls. Bars are SE.

Susceptibility to activation-induced cell death after T-cell-receptor engagement Since preactivated T cells that express CD95 (Fas) are susceptible to activation-induced cell death after stimulation via the T-cell-receptor complex, we investigated whether the observed defects in the proliferative responses of T cells in LVAD recipients after T-cell-receptor engagement might be related to activation-induced cell death. Figure 6 is an example of the flow cytometry assay used to detect the proportion of CD4 T cells in an LVAD recipient and a control that underwent apoptosis and cell death after 24 h of culture with medium or monoclonal antibodies to CD3. In the representative LVAD recipient in figure 6, 72% of resting CD4 T cells expressed phosphatidylserine, which is consistent with cells in the early phases of apoptosis. After 24 h of culture in medium, 7% of annexin-V-binding CD4 T cells underwent cell death. After 24 h of culture with monoclonal antibodies to CD3, the proportion of annexin-V-binding CD4 T cells that underwent cell death increased by three-fold to 21%. By contrast, in the representative control in figure 6, there was no difference in the proportion of annexinV-binding cells that underwent cell death after culture with medium or monoclonal antibodies to CD3 (5% to 7%). The increase in CD4 T-cell death after activation with antibodies to CD3 was then compared in LVAD recipients and controls. After activation of resting T cells with monoclonal antibodies to CD3, the proportion of annexin-V-binding CD4 T cells that underwent cell death increased by a mean of 3·2-fold in LVAD recipients, compared with only 1·2-fold in controls (p<0·05; figure 7). Since these results show that circulating CD4 T cells from LVAD recipients have increased susceptibility to activation-induced cell death compared with those from controls, we measured T cells in a cross-sectional analysis of LVAD recipients and controls. The mean number of circulating CD4 T cells was significantly lower in LVAD recipients than in controls (374 [6] vs 624 [59] per ␮L, p<0·01). By contrast, mean CD8 T-cell counts did not differ significantly between the

groups (230 [10] vs 194 [32] per ␮L, p<0·05). These observations indicate that the increased susceptiblity to activation-induced cell death in LVAD recipients may result in a reduction in the number of circulating CD4 T cells.

Discussion We found that LVAD implantation is accompanied by progressive defects in cellular immunity that seem to be the result of an aberrant state of T-cell activation involving the CD95 (Fas) pathway and activationinduced cell death of CD4 T cells. These defects predispose LVAD recipients to candidal and other systemic infections,17–19 and underlines the importance of the host-graft interaction as an integral part of LVAD biology beyond its mechanical function as a lifesustaining pump. Since the T-cell defects are accompanied by excessive B-cell reactivity, including production of antibodies against HLA molecules and phospholipids,27,28 this interaction seems to result in a loss of T cells involved in cellular immunity and immunoregulatory processes. Our findings point to an underlying immunological mechanism that accounts for the high rate of infectious complications in LVAD recipients. At the time of explantation, the LVAD surface is extensively coated with cells of monocyte or dendritic lineage26 that are functionally activated, as defined by expression of nuclear factor kB29 and augmented production of cytokines and coagulation factors.30 Excessive delivery of costimulatory signals by the activated antigenpresenting cells on the LVAD surface to interspersed T cells presumably accounts for increased expression of markers of T-cell activation, such as interleukin-2receptors 26 and CD95. Since triggering via the T-cellreceptor complex increases expression of CD95(Fas) ligand (CD95L),31 subsequent T-cell exposure to antigenic stimulation may result in activation-induced cell death, rather than the appropriate response of proliferation, because of interactions between CD95 and newly expressed CD95L. The coexistence of defects in T-cell immunity and prominent B-cell hyper-reactivity in LVAD recipients is similar to another disorder of the immune system, infection with HIV-1.32 Progressive depletion of CD4 T cells and immune dysfunction in HIV-1 infection accompanies progressive increases in viral burden within antigen-presenting cells such as macrophages and dendritic cells.33–35 One proposed mechanism is inappropriate induction of apoptotic T-cell death, resulting from HIV-1-mediated interactions between CD95 (Fas) and CD95L (FasL),36–38 suggesting the involvement of excessive T-cell costimulation by HIV-1 infected antigen-presenting cells. However, by contrast with LVAD implantation, the severe immune dysfunction associated with HIV-1 infection is multifactorial, and apoptosis may be one of many defects induced by infection with the virus. Since mechanical support is under investigation as a permanent therapy for heart failure, the immune dysfunction identified in this report will need to be addressed for successful long-term LVAD implantation. One potential approach to prevent activation-induced cell death is the use of cyclosporin A or tacrolimus (FK 506), two drugs that inhibit mRNA transcription of CD95L after activation of T cells via the T-cell-receptor

complex.39 We are currently examining these and other approaches to reduce the abnormal immune activation in LVAD recipients. Contributors Hendrik Jan Ankersmit gathered and analysed the clinical data, and designed and performed experimental studies. Sorina Tugulea helped with laboratory investigations. Talia Spanier and John Artrip helped collect and analyse clinical data. Elizabeth Burke and Margaret Flannery were responsible for execution of the study, including acquisition and management of clinical and laboratory data. Alan Weinberg did the statistical analyses. Donna Mancini and Mehmet Oz supervised all clinical activities. Niloo Edwards and Eric Rose provided infrastructure support. Silviu Itescu designed and coordinated the study, analysed the data, and oversaw the writing and editing of the paper.

References 1

2

3 4

5

6

7

8 9

10

11

12

13

14

15

16

17

18

19

20

21

Schocken DD, Arrieta MI, Leaverton PE, Ross EA. Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol 1992; 20: 301–06. Effects of enalapril on mortality in severe congestive heart failure: results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS) Trial Study Group. N Engl J Med 1987; 316: 1429–35. Hognes JR. The artificial heart: prototypes, policies and patients. Washington, DC: National Academy Press, 1991: 1–312. Annual Report of the US Scientific Registry for Organ Transplantation and the Organ Procurement and Transplantation Network, 1990. Washington, DC: US Department of Health and Human Services, 1990. Frazier OH, Rose EA, Macmanus Q, et al. Multicenter clinical evaluation of the Heartmate 1000 IP left ventricular assist device. Ann Thorac Surg 1992; 102: 578–87. McCarthy PM, Rose EA, Macmanus Q, et al. Clinical experience with the Novacor ventricular assist system. J Thorac Cardiovasc Surg 1991; 102: 578–87. Oz MC, Argenziano M, Catanese KA, et al. Bridge experience with long-term implantable left ventricular assist devices: are they an alternative to transplantation? Circulation 1997; 95: 1844–52. Didisheim P. Current concepts of thrombosis and infection in artificial organs. ASAIO J 1994; 40: 230–37. Wagner WR, Johnson PC, Kormos RL, Griffith BP. Evaluation of bioprosthetic valve-associated thrombus in ventricular assist device patients. Circulation 1993; 88: 2023–29. Rose EA, Levin HR, Oz MC, et al. Artificial circulatory support with textured interior surfaces: a counterintuitive approach to minimizing thromboembolism. Circulation 1994; 50: 1187–91. Menconi MJ, Owen T, Dasse KA, Stein G, Lian JB. Molecular approaches to the characterization of cell and blood/biomaterial interactions. J Card Surg 1992; 7: 177–87. Dasse KA, Chipman SD, Sherman CN, Levine AH, Frazier OH. Clinical experience with textured blood contacting surfaces in ventricular assist devices. ASAIO Trans 1987; 33: 418–25. Holmann WL, Murrah CP, Ferguson ER, Bourge RC, McGriffin DC, Kurklin JK. Infections during extended circulatory support: University of Alabama at Birmingham experience 1989 to 1994. Ann Thorac Surg 1996; 61: 366–71. Hermann M, Weyand M, Greshake B, et al. Left ventricular assist device infection is associated with increased mortality but is not a contraindication to transplantation. Circulation 1997; 95: 814–17. Goldstein D, El-Amir NG, Ashton RC, et al. Fungal infections in left ventricular assist device recipients—incidence, prophylaxis and treatment. ASAIO Trans 1995; 41: 873–75. McCarthy PM, Schmitt SK, Vargo RL, Gordon S, Keys TF, Hobbs RE. Implantable LVAD infections: implications for permanent use of the device. Ann Thorac Surg 1996; 61: 359–65. Argenziano M, Cantanese KA, Moazimi N, et al. The influence of infection on survival and successful transplantation in patients with left ventricular assist devices. J Heart Lung Transplant 1997; 16: 822–31. Fischer SA, Trenholme GM, Costanzo MR, Piccione W. Infectious complications in left ventricular assist device recipients. Clin Infect Dis 1997; 24: 18–23. Weyand M, Hermann M, Kondruweit M, et al. Clinical impact of infections in left ventricular assist device recipients: the importance of site and organism. Transpalnt Proc 1997; 29: 3327–29. McBride LR, Ruzevich SA, Pennington DG, et al. Infectious complications associated with ventricular assist device support. ASAIO Trans 1987; 33: 201–02. Robins JP. Current protocols in cytometry vols 1, 2. New York: John Wiley & Sons, 1997.

22 Koopman G, Reutelingsperger CP, Kuiten GA, Keehnen RM, Pals ST, van Oers MH. Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis. Blood 1994; 84: 1415–20. 23 Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled annexin V. J Immunol Meth 1995; 184: 39–51. 24 Tilly JL, Hsueh AJW. Microscale autoradiographic method for the qualitative and quantitative analysis of apoptotic DNA fragmentation. J Cellular Physiol 1993; 154: 519–26. 25 Michalopoulos A, Stavridis G, Geroulanos S. Severe sepsis in cardiac surgical patients. Eur J Surg 1998; 164: 217–22. 26 Spanier TB, Rose S, Schmidt AM, Itescu S. Interactions between dentritic cells and T cells on the surface of left ventricular assist devices leads to a TH2 pattern of cytokine production and B cell hyperreactivity in vivo. Circulation 1996; 94 (suppl 8): 1704. 27 Massad MG, Cook DJ, Schmitt SK, et al. Factors influencing HLA sensitization in implantable LVAD recipients. Ann Thorac Surg 1997; 64: 1120–25. 28 Itescu S, Tung TC, Burke EM, et al. Preformed IgG antibodies against non donor-specific MHC class II antigens are major risk factors for high-grade cellular rejection in recipients of cardiac transplantation. Circulation 1998; 98: 786–93. 29 Spanier TB, Oz MC, Rose EA, et al. Activation of NF-KB is central to the proinflamatory/procoagulant response in textured surface left ventricular assist device recipients and may be influenced by antiinflammatory intervention with aspirin. J Heart Lung Transpl 1997; 17: 80.

30 Spanier T, Oz M, Levin H, et al. Activation of coagulation and fibrinolytic pathways in patients with left ventricular assist devices. J Thorac Cardiovasc Surg 1996; 112: 1090–97. 31 Ju ST, Panka DJ, Cui H, et al. Fas(CD95)/FasL interactions required for programmed cell death after T-cell activation. Nature 1995; 373: 444–48. 32 Weiss RA. How does HIV cause AIDS. Science 1993; 260: 1273–79. 33 Pantaleo G, Graziosi C, Demarest JF, et al. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 1993; 362: 355–58. 34 Perelson AS, Neumann AU, Markowitz M, Leonard JM, Ho DD. HIV-1 dynamics in vivo: virion clearance rate, infected cell life-span, and viral generation time. Science 1996; 271: 1582–86. 35 Safrit JT, Koup RA. The immunology of primary HIV infection: which immune responses control HIV replication. Curr Opin Immunol 1995; 7: 456–61. 36 Ameisen JC, Capron A. Cell dysfunction and depletion in AIDS: the programmed cell death hypothesis. Immunol Today 1991; 12: 102–05. 37 Groux H, Torpier G, Monte D, Mouton Y, Capron A, Ameisen JC. Activation-induced death by apoptosis in CD4+ T cells from human immunodeficiency virus infected asymptomatic individuals. J Exp Med 1992; 175: 331–40. 38 Katsikis PD, Wunderlich ES, Smith CA, Herzenberg LA, Herzenberg LA. Fas antigen stimulation induces marked apoptosis of T lymphocytes in human immunodeficiency virus-infected individuals. J Exp Med 1995; 181: 2029–36. 39 Brunner T, Yoo NJ, LaFace D, Ware CF, Green DR. Activationinduced cell death in murine T cell hybridomas: differential regulation of Fas (CD95) versus Fas ligand expression by cyclosporin A and FK 506. Int Immunol 1996; 8: 11017–19.