|Year : 2021 | Volume
| Issue : 3 | Page : 142-146
A study of prothrombotic factors in human immunodeficiency virus-infected patients
Meena Lanjiwar1, Pushpa Kumari2, Jitendra Mohan Khunger3, Narender Singh Negi2
1 Department of Neurology, All India Institute of Medical Sciences, New Delhi, India
2 Department of Medicine, VMMC and Safdarjung Hospital, New Delhi, India
3 Department of Hematology, VMMC and Safdarjung Hospital, New Delhi, India
|Date of Submission||18-Oct-2020|
|Date of Decision||26-Oct-2020|
|Date of Acceptance||27-Oct-2020|
|Date of Web Publication||09-Jul-2021|
Dr. Jitendra Mohan Khunger
Consultant Haematologist and Assoc. Professor, Department of Haematology, Vardhman Mahavir Medical College and Safdar Jang Hospital, New Delhi - 110 029
Source of Support: None, Conflict of Interest: None
Introduction: Human immunodeficiency virus (HIV) infection has been described as a hypercoagulable state. Some studies suggest that thrombotic events may be 2–10 times more prevalent in this group than in the general population. Subjects and Methods: This was a single-center, cross-sectional case–control study. Thirty confirmed HIV cases of age group 20–60 years were enrolled. A similar number of age- and sex-matched healthy volunteers were taken as controls. CD4 cell count, protein C, protein S, antithrombin III (AT III), anticardiolipin antibody, and highly sensitivity C-reactive protein (hs-CRP) estimation was done and correlated with the CD4 counts. Results: Among HIV cases, 40% were protein C deficient, while 10% were deficient in control group (P = 0.015). One-third (33%) were AT III deficient in HIV-infected group, while none were observed AT III deficient in control group (P < 0.001). Protein C and antithrombin levels did not correlate with CD4 count. Protein S levels were normal in all the HIV-infected cases. Anticardiolipin antibody was not detected in any patient among HIV group. Mean hs-CRP levels in cases and controls were 20.18 ± 12 mg/dl and 3.32 ± 5.13 mg/dl, respectively (P < 0.0005). Conclusion: Protein C deficiency and AT III deficiency were found to be significantly associated with HIV infection. The hs-CRP levels were significantly elevated in HIV-infected patients' protein C deficiency, increased level of hs-CRP, and AT III deficiency did not correlate with CD4 cell count.
Keywords: CD4 count, human immunodeficiency virus, hypercoagulable, thrombosis
|How to cite this article:|
Lanjiwar M, Kumari P, Khunger JM, Negi NS. A study of prothrombotic factors in human immunodeficiency virus-infected patients. Indian J Med Spec 2021;12:142-6
|How to cite this URL:|
Lanjiwar M, Kumari P, Khunger JM, Negi NS. A study of prothrombotic factors in human immunodeficiency virus-infected patients. Indian J Med Spec [serial online] 2021 [cited 2021 Sep 19];12:142-6. Available from: http://www.ijms.in/text.asp?2021/12/3/142/321048
| Introduction|| |
Human immunodeficiency virus (HIV) infection is a global pandemic, with approximately 38 million adults infected worldwide. India has the third highest number of people living with HIV in the world, with 2.1 million Indians.
Increase in survival with newer highly active antiretroviral therapy (HAART) has brought into notice many more complications during the course of the disease. Improved survival has been followed by an anticipated and increased prevalence of non-AIDS-related conditions, in particular cardiovascular diseases and thromboembolic events, which are now a leading cause of morbidity and mortality among HIV-infected people.
HIV infection has been described as a hypercoagulable state in the literature that predisposes to the development of serious and potentially life-threatening thromboembolic disorders such as deep vein thrombosis, pulmonary embolism, hepatic vein thrombosis, and inferior vena cava thrombosis. Some studies suggest that thrombotic events may be 2–10 times more prevalent in this group than in the general population.,, Autopsy studies reveal high rates of previously undiagnosed thromboembolism among patients with HIV.
Researchers noted that thrombosis was more common in patients with opportunistic illness, indinavir users, and hospitalized and older population. Abnormalities thought to be predisposing to a hypercoagulable state in HIV infection include the presence of procoagulants such as antiphospholipid antibodies and lupus anticoagulant, increased platelet activation, elevated homocysteine levels, acquired protein S deficiency, deficiency of protein C, heparin cofactor II and antithrombin III (AT III), and increased levels of Von Willebrand factor, highly sensitivity C-reactive protein (hs-CRP), and d-dimers.
Thromboembolic phenomena are potentially preventable, and it is of utmost importance to identify individuals at high risk so that they may benefit from primary thromboprophylaxis. However, little work has been done on defining the exact mechanisms by which thrombophilia occurs in HIV disease. There are even fewer studies in Indian subset of patients. This study was undertaken to study the levels of different procoagulants in asymptomatic HIV-infected patients and its correlation with the degree of HIV-induced immunosuppression as determined by CD4 cell counts.
| Subjects and Methods|| |
This was a single-center, cross-sectional case–control study. A total of thirty confirmed HIV cases as per NACO guidelines of either sex attending the antiretroviral therapy clinic of age group of 20–60 years were enrolled in the study. A similar number of age- and sex-matched healthy volunteers were taken as controls. Patients on anticoagulant therapy, active tuberculosis (TB), history of hypertension, diabetes mellitus, pregnancy and postpartum <6 months, oral contraceptive pill use, postsurgical <6 months, any underlying malignancy, opportunistic infections, and autoimmune disorders were excluded from the study. Approval from the institutional ethical committee and a written informed consent from all the subjects enrolled for the study were taken.
All the recruited patients underwent the following investigations: complete hemogram, random blood sugar, liver and kidney function test, serum electrolytes, urine routine and microscopy, chest X-ray posteroanterior view, protein C, protein S, AT III, anticardiolipin antibody (ACA), and hs-CRP.
Protein Cestimation was done by Reaads® Protein C Antigen Test Kit for in vitro Diagnostic Use. Protein C antigen values were expressed in relative percentage (%) as compared to pooled normal plasma. Normal range of protein C antigen for this assay was 72%–160% (mean 110%, standard deviation [SD]: 24). Protein S estimation was done by Reaads® Protein S Antigen test kit for in vitro Diagnostic Use. Total and free protein S values were expressed in relative percentage (%) as compared to normal pooled plasma. Normal range for total and free protein S for this assay was 60%–150% and 50%–130%, respectively. AT III estimation was done by Biocientifica SA Diffu-Plate. Normal range was 5.9–78.8 mg/dl.
ACA test was done by Calbiotech anti-Cardiolipin immunoglobulin G (IgG) enzyme-linked immunosorbent assay (ELISA) kit. Levels were labeled as undetectable (<0.9), borderline positive (0.9–1.1), and detectable (>1.1).
hs-CRP estimation was done by The EiAsy™ Way Diagnostic Biochem Canada Inc. Cat no. CAN- CRP-4360 Version: 4.0 h-CRP ELISA Kit. Levels were estimated as low, intermediate, and high risk for values of <1.0 mg/L, 1.0–3.0 mg/L, and >3.0 mg/L, respectively.
All the statistical analyses were carried out using SPSS (Statistical Package for the Social Sciences, IBM Inc., Chicago, Illinois, USA) Version 21.0. Categorical variables were presented in number and percentage (%) and continuous variables were presented as mean ± SD and median. The evaluation of the diagnostic parameters, i.e., of protein C, protein S, AT III, hs-CRP, and anticardiolipin antibodies was in terms of quantitative levels. Data were expressed as mean, SD, median, or percentage. Student's t-test was applied on quantitative data and statistical difference of normal variables among different group was compared. Independent t-test or Mann–Whitney test was used for skewed variable. P ≤ 0.05 was taken as the level of statistical significance.
| Results|| |
Thirty-eight subjects were screened, and thirty were finally included in the study. The most frequent causes of exclusion were active TB, opportunistic infection, AIDS-related malignancy, and the use of protease inhibitors. Thirty healthy normal subjects were taken as controls. The minimum and maximum duration of disease was 2 years and 8 years, respectively, average duration of disease being 4.5 years. The majority of the HIV cases were in 20–40 years age group (n = 25, 83.33%). Among controls, 80% (n = 24) were in this age group. In HIV group, 66.67% (n = 20) were males and 33.33% (n = 10) were female. Whereas in control group, 80% (n = 24) were male and 20% (n = 6) were female. The two groups were comparable in terms of basic demographic parameters. Twenty-eight patients were on zidovudine, lamivudine, and nevirapine, whereas two were on tenofovir, lamivudine, and nevirapine.
Highest and lowest CD4 cell count was 687 cell/mm3 and 66 cell/mm3 with mean of 322.77 ± 150.86 cell/mm3 among HIV-infected group. Among these patients, 16.67% (n = 5) patients had CD4 cell count <200 cell/mm3, 70% (n = 21) had between 200 and 500 cell/mm3, and 13.33% (n = 4) had more than 500 cell/mm3.
Forty percent (n = 12) of patients in HIV group, whereas 10% (n = 3) healthy controls were protein C deficient (P = 0.015).With reference range of 72%–160%, mean value of protein C levels among cases was 0.79 ± 0.22 with minimum of 52% to maximum 129%. Cases had lesser value of protein C as compared to controls (P < 0.0004) [Table 1].
|Table 1: Protein S, protein C, antithrombin III and high sensitivity C reactive protein levels in cases versus control|
Click here to view
One-third (33.33%) of protein C deficient patients were young (21–30 years of age), whereas only one case was in 61–70-year age group (P = 0.402). Among cases with CD4 cell count <200 cell/mm3 (n = 5), 40% (n = 2) were protein C deficient. Nearly 42.48% (n = 9) were observed protein C deficient among cases with CD4 cell count between 200 and 500 cell/mm3 (n = 21). While only 25% (n = 1) with CD4 cell count more than 500 cell/mm3 was found to be protein C deficient. Protein C deficiency was associated with HIV infection irrespective of different CD4 cell count levels (P = 0.800).
With reference range of protein S (60%–150%), cases had a mean value of 0.94 ± 0.18 as compared to controls with a mean value of 1.21 ± 0.28. The maximum protein S level among cases was 135% with a minimum of 56%, as compared to controls (min-max, 71%–213%). Levels of protein S in HIV-infected cases were in the normal reference range; however, cases had a lesser value of protein S than that of controls (P < 0.0005) [Table 1].
It was observed that 33.33% (n = 10) were AT III deficient in HIV-infected group, while no one was found to be AT III deficient in control group (P = 0.001). AT III deficiency was independently associated with HIV infection and did not correlate with CD4 cell count
(P = 0.691). ACA was not detected in any patient among HIV group.
With reference range of hs-CRP 1–3 mg/dl, cases had mean value of 20.18 ± 12.25 as compared to controls (mean 3.32 ± 5.13) (P < 0.0005) [Table 1]. Maximum hs-CRP level among cases was 34 mg/dl, whereas among controls it was 21.5 mg/dl.
The hs-CRP was observed to be elevated (>3 mg/L) in 27 cases (90%) as compared to eight controls (26.67%). No association was observed between hs-CRP levels and CD4 cell count (P = 0.69).
All AT III deficient cases had hs-CRP levels >3 mg/dl, suggesting linear relationship between AT III deficiency and HIV-induced inflammatory and hypercoagulable state, resulting in an increased risk of thromboembolic events in these patients. Among 12 protein C deficient cases, majority (n = 11) had hs-CRP levels >3 mg/dl.
| Discussion|| |
Even though HIV is recognized as a prothrombotic state and a risk factor for venous and arterial thrombosis, there is little evidence regarding the mechanism. This study was undertaken to evaluate the levels of different procoagulants in HIV-infected patients and its co-relationship with the degree of immunosuppression.
We found an incidence of 40% of acquired protein C deficiency among HIV-infected patients with statistically significant association (P = 0.015). Our finding reconfirms the observation of Orlovic and Smego a retrospective study of 12 HIV seropositive patients, where three out of 12 HIV patients had protein C deficiency. Similar findings of increased incidence of protein C deficiency among HIV-infected patients were noted by Saif et al., in their retrospective study of 131 HIV-infected patients.
Erbe et al. in their study of thrombophilic parameters in 49 HIV-infected patients with acute opportunistic infections found an increased incidence of protein C deficiency during acute settings (25%, 12/49), with improvement in protein C levels after successful treatment on follow-up visit after 10 weeks, with incidence dropping to 8% (2/26). We, however, excluded patients with acute illness and opportunistic infections from the study.
The proposed mechanisms for acquired protein C deficiency in HIV are multifactorial, including altered synthesis and metabolism of protein C, as well as low-grade disseminated intravascular coagulation (DIC) with consumptive coagulopathy, in the setting of HIV infection with severe immunosuppression. Antiphospholipid antibodies have also been postulated to interfere with the activated protein C/protein S complex at the phospholipid surface and cause inhibition of the activated protein C complex. A link between opportunistic infections and thrombosis through endothelial activation, has been suggested, through upregulation of some cytokines such as tumor necrosis-factor alpha (TNFα), interleukin (IL-1), IL-6, factor VIII, and fibrinogen, as well as downregulation of fibrinolytic proteins, protein C (consumption as an anti-inflammatory mediator).
Our study did not find an association between protein C deficiency and CD4 counts; thus, its association with HIV infection is independent irrespective of severity of disease and virus-mediated immunosuppresion. On the contrary, Feffer et al. established a direct correlation between decreased CD4 cell count and decreased protein C activity, with patients having CD4 cell count of <200 cells/mm3 having significantly less protein C activity than those with > 400 cells/mm3 (P < 0.005). This evidence supports the theory that functional protein C declines with immunosuppression, thus increasing the risk of venous thromboembolism (VTE).
We did not find any protein S deficient cases among HIV-infected group. Although we found lesser mean value of protein S in HIV-infected patient (0.94 ± 0.18 vs. 1.21 ± 0.28,
P < 0.0005). This finding is in contrast with the finding of most of the earlier western studies where protein S deficiency was consistently found to be associated with HIV infection. In fact, protein S deficiency has been found to be the most common and consistent coagulation abnormality in HIV with reported prevalence ranging from 33% to 94%.
Decreased synthesis of protein S by the endothelial cells, hepatocytes, and megakaryocytes injured in HIV infection, antibodies to protein S and low levels of circulating free antigen, TNF α can lower the levels of active protein S, and loss of protein S in urine in HIV-related nephropathy has been postulated as mechanisms of protein S deficiency in HIV infection.
Bissuel et al. in their study found free protein S levels were significantly lower in patients with full-blown AIDS than in patients without AIDS, suggesting possible association with severity of disease and degree of immunosuppression. In a follow-up investigation performed by Erbe et al., among 41 acutely ill HIV-infected patients, protein S levels increased significantly from initial to follow-up visit (P < 0.05), suggesting that an acquired protein S deficiency often develop in patients with HIV during acute illness mainly because of increase in C4-binding protein (acute phase protein), resulting in transient decrease in free levels of protein S which may be reversible after treatment for opportunistic infections.
C4b-Bp is an acute phase reactant and increases with inflammation. This can lead to a temporary reduction in free protein S antigen and protein S activity. Boullanger et al., however, did not observe a decrease of free protein S in inflammatory syndromes other than HIV even though C4b-BP was increased. They demonstrated that a higher level of C4b-binding protein did not reduce the level of free protein S, suggesting protein S deficiency could result from endothelial cell dysfunction during HIV infection per se.
Although the reason of having normal protein S level among HIV infected patient in this study as compared to earlier studies in western world cannot be fully explained, some of these earlier studies were retrospective, usually enrolled randomly selected HIV-infected patients resulting in variable numbers of subjects with severe immune suppression (CD4 cell count <200 cell/mm3) and opportunistic infection, whereas others already had thrombosis and AIDS-defining illnesses. These factors had already been excluded from our study during enrolment of the cases. Moreover, racial variation and prevalence of protein S deficiency in Indian population also could not be excluded, as clear data regarding the prevalence of protein S deficiency are lacking in Indian population. Evaluation of several other inflammatory markers (e.g., TNF α, IL-1), antibodies against protein S, levels of C4-binding protein, and lupus anticoagulant could have been useful in explaining the normal protein S levels seen in the study.
Ten cases (33.33%) were found AT III deficient in HIV-infected group (P = 0.001). There is no direct evidence of HIV infection leading to antithrombin deficiency. Acquired AT III deficiency frequently occurs in the course of HIV disease as a consequence of associated conditions that cause decreased protein synthesis (liver diseases e.g., hepatitis C virus coinfections, and malnutrition), protein-losing nephropathies or enteropathies, and consumptive states (malignancy, DIC, surgery).
Of the thirty HIV patient enrolled in the study, none were found to be positive for ACA. Eighty-four HIV-infected patients were studied by Daroca et al. Of the 84 HIV-positive patients, 50 were IgG-ACA positive (59.5%) and 34 IgG-ACA negative (40.5%). No significant differences were found with stage of disease with IgG-ACA positivity in
HIV-positive patients. ACA does not appear to be a prognostic marker in HIV-infected subjects. The presence of IgG-ACA is probably related to HIV infection itself and is indicative of impaired humoral immunity in these patients.
Saif et al. concluded that HIV infection induces the destruction of CD4 lymphocytes, which leads to polyclonal stimulation of B cells and hypergammaglobulinemia, resulting in antibodies against damaged endothelial cells (exposed phospholipids) and inhibition of protein S synthesis. It is thought that lupus anticoagulant activity in these patients might be an epiphenomenon secondary to chronic immune stimulation in HIV infection and no pathogenic correlation has been found with thromboses. Furthermore, autoantibodies to phospholipids, including lupus anticoagulant and anticardiolipin, transiently appear in a large proportion of HIV-infected patients.
Negative ACA and normal protein S level in our study could be due to the reason that all enrolled subjects were asymptomatic and most of the patients (n = 21) had CD4 cell count between 200 and 500 cells/mm3, suggesting lesser destruction of CD4 lymphocytes and thereby lesser polyclonal stimulation of B cells and hypergammaglobulinemia, resulting in lesser antibodies against damaged endothelial cells (exposed phospholipids) and inhibition of protein S synthesis.
Most of the cases (n = 27) had hs-CRP level more than 3 mg/L and the levels were significantly elevated (P < 0.0005). This indicates the possible role of higher levels of hs-CRP as pro-inflammatory marker in pathophysiology of hypercoagulable state in HIV infection. Statistical analysis was, however, suggestive of insignificant association between different hs-CRP levels and CD4 cell count (P = 0.691), supporting the fact that HIV infection is a pro-inflammatory state per se, promoting inflammatory cascade in body irrespective of CD4 cell count. Among 12 protein C-deficient cases, 11 had hs-CRP levels >3 mg/dl. All the AT III-deficient cases had hs-CRP levels >3 mg/dl, suggesting that linear relationship between AT III deficiency and HIV-induced inflammatory and hypercoagulable state exist, resulting in increased risk of thromboembolic events in these patients.
In a previous study, Triant et al. concluded that elevated CRP and HIV are independently associated with increased acute myocardial infarction (MI) risk, and patients with HIV and increased CRP have a markedly increased relative risk of acute MI. Measurement of CRP may be useful in the cardiovascular risk assessment of patients of HIV.
Even though HIV disease was recognized as a prothrombotic condition around 20 years back, there is little evidence regarding the mechanisms causing it. With increase in life span of patients on HAART, patients now frequently present with manifestations other than opportunistic infections. Thromboembolism can present as both HIV and non-HIV-associated disorders and can be misdiagnosed, leading to delayed treatment. VTE can mimic opportunistic lung infections in patients with HIV; therefore, the former should be considered in differential diagnosis of lung diseases in this setting, when patients have unexplained dyspnea or hypoxemia. Physicians should be aware of the high risk of thromboembolism in HIV patients (2–10 times more than normal population). Early suspicion of thrombosis or thromboembolism as differentia diagnosis may result in early diagnosis and treatment or prevention of thromboembolic events in these set of patients.
There were some limitations to the study. An important limitation was the small number of subjects included in the study. Normal levels of free protein S among HIV-infected cases were in contrast to the various previous studies. Due to some practical constraints, various other factors such as heparin cofactor II, factor V Leiden, Von-Willebrand factor, and homocysteine could not be done.
| Conclusion|| |
We found decreased protein C and AT III levels associated with HIV infection irrespective of CD4 count. Protein S levels were in normal range, but overall levels were lower. We also found significantly higher levels of hs-CRP in HIV-infected cases. Further studies are required to evaluate if the risk factors which are traditionally associated with increased VTE risk have the same impact in HIV patients also. More research is needed to comment if routine screening should be done in HIV-positive patients for procoagulant state and if the traditional VTE risk management strategies are applicable to HIV patients also.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Palella FJ Jr., Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338:853-60.
Orlovic D, Smego A. Hypercoagulability due to protein S deficiency in HIV seropositive patients. Int J Collab Res Intern Med Public Health 2009;1:187-93.
Saber AA, Aboolian A, LaRaja RD, Baron H, Hanna K. HIV/AIDS and the risk of deep vein thrombosis: A study of 45 patients with lower extremity involvement. Am Surg 2001;67:645-7.
Copur AS, Smith PR, Gomez V, Bergman M, Homel P. HIV infection is a risk factor for venous thromboembolism. AIDS Patient Care STDS 2002;16:205-9.
Sullivan PS, Dworkin MS, Jones JL, Hooper WC. Epidemiology of thrombosis in HIV-infected individuals. The Adult/Adolescent Spectrum of HIV Disease Project. AIDS 2000;14:321-4.
Wilkes MS, Fortin AH, Felix JC, Godwin TA, Thompson WG. Value of necropsy in acquired immunodeficiency syndrome. Lancet 1988;2:85-8.
Yu-Min PS, Eugene PF. Thrombosis and a Hypercoagulable State in HIV-Infected Patients. Clin Appl Thromb Hemost 2004;10:277-80.
Saif MW, Bona R, Greenberg B. AIDS and thrombosis. Retrospective study of 131 HIV-infected patients. AIDS Patient Care STDS 2001;15:311-20.
Erbe M, Rickerts V, Bauersachs RM, Lindhoff-Last E. Acquired protein C and protein S deficiency in HIV-infected patients. Clin Appl Thromb Hemost 2003;9:325-31.
Saif MW, Greenberg B. HIV and thrombosis: A review. AIDS Patient Care STDS 2001;15:15-24.
Heit JA, Kobbervig CE. Relationship between progression to AIDS and thrombophilic abnormalities in HIV Infection. Clin Chem 2008;54:1226-33.
Jacobson MC, Dezube BJ, Aboulafia DM. Thrombotic complications in patients infected with HIV in the era of highly active antiretroviral therapy: A case series. Clin Infect Dis 2004;39:1214-22.
Lafeuillade A, Alessi MC, Poizot-Martin I, Boyer-Neumann C, Zandotti C, Quilichini R, et al
. Endothelial cell dysfunction in HIV infection. J Acquir Immune Defic Syndr 1992;5:484-9.
Feffer SE, Fox RL, Orsen MM, Harjai KJ, Glatt AE. Thrombotic tendencies and correlation with clinical status in patients infected with HIV. South Med J 1995;88:1126-30.
Gladson CL, Scharrer I, Hach V, Beck KH, Griffin JH. The frequency of type I heterozygous protein S and protein C deficiency in 141 unrelated young patients with venous thrombosis. Thromb Haemost 1988;59:18-22.
Friedman KD, Marlar RA, Gill JC, Endres-Brooks J, Montgomery RR. Protein S deficiency in patients with lupus anticoagulant. Blood 1986;68:333.
Bissuel F, Berruyer M, Causse X, Dechavanne M, Trepo C. Acquired protein S deficiency: Correlation with advanced disease in HIV-1-infected patients. J Acquir Immune Defic Syndr (1988) 1992;5:484-9.
Boullanger N, El Kouri D, Trossaert M, Truchaud F, Trewick D, Planchon B. Coagulation activation in patients with an inflammatory syndrome: Is there a link with acquired protein S deficiency? Blood Coag Fibrinol 1998;9:167.
Daroca JC, Gutiérrez-Cabollada J, Yazbeck H, Berges A, Rubies-Prat J. Anticardiolipin antibodies and acquired immunodeficiency syndrome: Prognostic marker or association with HIV infection? Infection 1992;20:140-2.
Espinosa G, Cervera R. Antiphospholipid syndrome: Frequency, maincause and risk factors of mortality. Nat Rev Rheaumatol 2010;6:296-300.
Triant VA, Lee H, Hadigan C, Grinspoon SK. Increased acute myocardial infarction rates and cardiovascular risk factors among patients with human immunodeficiency virus disease. J Clin Endocrinol Metab 2007;92:2506-12.
Jacobson MC, Dezube BJ, Aboulafia DM. Thrombotic complications in patients infected with HIV in the Era of highly active antiretroviral therapy: A case series. Clin Infect Dis 2004;39:1222.