From the onset of the human immunodeficiency virus (HIV) pandemic in the 1980s to the recent coronavirus disease 2019 (COVID-19) pandemic, multiple viral pandemics have occurred and all have been associated with hematologic complications of varying severity.
To review the hematologic complications associated with the HIV and other viral pandemics, the current theories regarding their causation, and the incidence and clinical impact of these complications on infected patients.
Peer-reviewed medical literature and the author's personal experience.
The HIV and other viral pandemics have been associated with a variety of hematologic complications that often cause significant morbidity and mortality in affected patients. HIV infection is associated with multiple hematologic disorders, many of which have a lower incidence in the era of highly active antiretroviral therapy but still represent a major clinical problem for HIV-infected patients. Our understanding of the pathogenesis of HIV-related hematologic complications, including HIV-associated lymphoproliferative disorders, has evolved in recent years. Other viral pandemics, including H1N1 influenza, severe acute respiratory syndrome (SARS) coronavirus, Middle East respiratory syndrome (MERS) coronavirus, and COVID-19, have also been associated with hematologic complications of varying severity. Our emerging understanding of the pathogenesis of the hematologic complications of HIV, COVID-19, and other viral pandemics may help in prevention, correct diagnosis, and treatment of these complications in current and future pandemics.
Since the beginning of the human immunodeficiency virus (HIV) pandemic in the 1980s, HIV infection has been well-known to be associated with a variety of hematologic complications, including hematopoietic defects, immune-mediated thrombocytopenia purpura (ITP), thrombotic thrombocytopenia purpura (TTP), thromboembolism, and multiple lymphoproliferative disorders.1–3 Although the introduction of highly active antiretroviral therapy (HAART) resulted in a decrease in the incidence of many of these disorders, each still represents a significant clinical problem for HIV-infected patients and the incidence of some has actually increased post HAART. Careful study for many years of each of these hematologic complications of HIV infection has led to an improved understanding of their pathogenesis and in some cases improved therapy for these disorders.
In the ensuing years, multiple additional viral pandemics have occurred, including H1N1 influenza, severe acute respiratory syndrome (SARS) coronavirus, Middle East respiratory syndrome (MERS) coronavirus, and coronavirus disease 2019 (COVID-19). Each of these respiratory viral pandemics has been associated with a number of hematologic complications that have significantly impacted the clinical outcome of infected patients.4–6 Studies performed about these complications in the earlier pandemics allowed for improved treatment and potential prevention of some of them. The COVID-19 pandemic has been associated with some of the same hematologic complications as the earlier pandemics, in some instances of even greater severity. Knowledge of the mechanisms of these complications in the earlier infections has helped with the rapid study and elucidation of the complications in the COVID-19 pandemic.
In this review, the hematologic disorders and complications associated with viral pandemics, starting with the HIV pandemic and ending with the ongoing COVID-19 pandemic, will be described and the latest concepts of these disorders will be presented. The hope is that improved understanding of the pathogenesis of these complications will help in prevention, correct diagnosis, and treatment of similar complications in current and future viral pandemics.
HIV Infection and Associated Hematologic Diseases
From the very beginning of the HIV pandemic, hematologic complications and associated hematologic disorders were recognized as serious consequences of HIV infection. In addition to the profound impact of HIV infection on the immune system and resulting opportunistic infections, abnormalities of hematopoiesis, immune-mediated and non–immune-mediated thrombocytopenia, coagulation abnormalities, and increased development of several lymphoproliferative disorders were also quickly recognized.1–3 One category of hematologic disease, namely development of certain high-grade lymphomas, became part of the case definition of acquired immunodeficiency syndrome (AIDS). Before HAART, these various hematologic complications often represented a substantial component of the clinical impact of HIV infection on the patient, and development of some of them was frequently the first clinical manifestation of HIV infection. Introduction of HAART around 1996 resulted in a dramatic improvement in the overall course of HIV infection and decreased the incidence of many, but not all, of these associated hematologic disorders.1–3 The nonneoplastic and neoplastic hematologic complications associated with HIV infection and the change in their incidence in the HAART era are listed in Tables 1 and 2.
Impact of HIV Infection on Hematopoiesis
Untreated HIV infection is often associated with anemia, neutropenia, and/or thrombocytopenia. These abnormalities may be secondary to other HIV-associated disorders and/or therapy for those disorders, including opportunistic infections, but HIV infection also clearly has a direct effect on hematopoiesis. The pathogenesis of the inhibition/disruption of hematopoiesis has been the subject of much investigation for many years, but is still poorly understood. The virus itself is known to infect cells that play a part in the control of hematopoiesis, such as macrophages and marrow stromal cells, which may disrupt production of hematopoietic growth factors and negatively impact the marrow microenvironment.3,7 In addition, HIV-related proteins induce aberrant production of inflammatory cytokines that may negatively impact hematopoiesis.3,8 Direct HIV infection of megakaryocytes was recognized early in the pandemic and clearly contributes to thrombocytopenia in some of these patients.9 Early studies failed to detect HIV infection of CD34+ hematopoietic progenitor cells, but later studies showed that these cells may be infected with HIV in at least a subset of patients.10 Bone marrow findings in untreated HIV-infected patients include significant morphologic evidence of dysplastic hematopoietic maturation as additional evidence of disrupted hematopoiesis.11 All of these findings strongly point to significant direct inhibition and disruption of hematopoiesis by HIV as a major cause of blood cell cytopenias in these patients. With the institution of HAART, HIV-associated cytopenias are frequently reversed,1,12 likely as a result of improvement in the various mechanisms responsible for the cytopenias; however, HAART itself appears to introduce additional abnormalities in the marrow microenvironment that may contribute to continued cytopenias in some patients.3
HIV-Related Immune-Mediated and Non–Immune-Mediated Thrombocytopenia
Immune-mediated thrombocytopenia, similar in many respects to conventional ITP, was one of the first manifestations of HIV infection recognized13 and remained a common clinical problem before the HAART era. Multiple mechanisms are thought to be responsible for the immune destruction of platelets in these patients, including immune complex-related destruction and anti–platelet-glycoprotein antibody production related to cross-reactivity between platelet glycoproteins and HIV.14 The incidence of this form of platelet destruction associated with severe thrombocytopenia is decreased by HAART, but less severe ITP continues to occur in patients treated with HAART, particularly in the subset of HIV-infected patients also infected with hepatitis C.15
Another important cause of thrombocytopenia in HIV-infected patients is TTP. HIV infection is associated with a 15- to 40-fold increase in TTP as compared to non–HIV-infected patients.16 The mechanism of HIV-related TTP is thought to be multifaceted, including cytokine-driven overproduction of von Willebrand factor (vWF), decreased production of the metalloprotease ADAMTS13, and production of anti-ADAMTS13 antibodies, all contributing to enhanced microangiopathic thrombosis and platelet consumption.17 Unlike with ITP-like thrombocytopenia, HAART has not significantly decreased the incidence of TTP in HIV-infected patients and it remains a major cause of morbidity and potential mortality.16
HIV-Related Coagulation Abnormalities
HIV infection is associated with significant coagulation abnormalities that tend to be worse with advanced immunosuppression and the associated secondary complications of HIV.18 The mechanisms of these abnormalities are multiple, including endothelial cell dysfunction, activation of coagulation factors, disrupted production of coagulation regulatory proteins, and production of lupus anticoagulant and other anti-phospholipid antibodies. Most HIV-associated coagulation abnormalities do not abate fully with HAART, and HAART itself may be associated with decreased coagulation factor synthesis and further endothelial cell dysfunction, both contributing to persisting coagulation defects even in treated HIV-infected patients.19
The coagulation abnormalities associated with HIV infection typically induce a hypercoagulable state in infected patients and associated increased incidence of thromboembolic complications. The incidence of clinical thrombotic events in HIV-infected patients varies in different patient populations but is reported to be as high as 8%.20 The coagulation abnormalities tend to be worse in patients with lower CD4 cell counts and a higher HIV viral load, but persist even when HIV infection is better controlled. Many of these abnormalities center on dysfunction of endothelial cells induced primarily by the impact of HIV glycoprotein 120 (gp120) and transactivator of transcription (Tat) on the activation or disruption of different endothelial cell functions, leading to activation of the coagulation cascade, increased cytokine production, and decreased production of tissue plasminogen activator and protein S.21 Overproduction of vWF and deficiency of protein C, antithrombin III, and heparin cofactor II induced by HIV infection also contribute to the hypercoagulable state.18
In addition to contributing to an increase in thromboembolic events, endothelial cell dysfunction and hyperactivation of the innate immune system induced by HIV are also thought to be the underlying cause of increased atherosclerosis and coronary artery disease in HIV-infected patients,22 complications that continue in spite of HAART. Increased morbidity and mortality in HIV-infected patients secondary to cardiovascular disease have been particular problems as these patients live longer with successful HAART.21
HIV-Associated Lymphoproliferative Disorders
A substantial increase in the incidence of certain high-grade lymphoproliferative disorders in HIV-infected patients has been recognized since the beginning of the pandemic and continues to be a significant clinical problem in these patients.2,23 This increase in incidence has been well documented to be the result of a combination of decreased immune surveillance, cytokine overproduction, chronic antigenic stimulation, and poor control of oncogenic viruses. More recent research has shown that HIV itself can directly induce lymphoma development, probably through the clonogenic effect of certain HIV proteins, such as p17 variants, on germinal center B lymphocytes.24,25
The HIV-related lymphoproliferative disorders are listed in Table 2. For many years, these disorders were thought to consist only of high-grade B-cell non-Hodgkin lymphomas (NHLs); however, based on more recent rethinking of the genesis of these disorders, lower-grade forms of lymphoproliferative disease are now included as well.23,26 Classic Hodgkin lymphoma (CHL) is also associated with HIV infection, but has never been accepted as an AIDS-defining disorder and its relationship to HIV has been less clear than with NHLs.27 Many of these disorders also occur in immunocompetent patients, whereas others are more specifically associated with HIV or other causes of immunodeficiency. Most HIV-related lymphomas, including both NHLs and CHL, are associated with Epstein-Barr virus (EBV) and/or human herpesvirus 8 (HHV8).
The incidence of many NHLs associated with HIV has decreased since the introduction of HAART (Table 2), but these disorders are still significantly more common in HIV-infected patients than the general population, with a HAART-era relative risk of nearly 10.28 The direct impact of long-term HIV infection on B lymphocytes may explain this continuing increased incidence even after HAART.24 Paradoxically, the incidence of CHL has significantly increased in HIV-infected patients in the HAART era compared to pre HAART.27
Diffuse Large B-Cell Lymphoma
Diffuse large B-cell lymphoma (DLBCL) is the most common type of HIV-related lymphoma, representing 45% to 65% of lymphomas in these patients in different populations. This lymphoma in HIV-infected patients is mostly EBV related and often exhibits an activated B-cell (ABC) immunophenotype and immunoblastic morphology. The ABC/immunoblastic type of HIV-related DLBCL is closely associated with severe immunodeficiency and a very high rate of EBV positivity (90% of cases) and has shown a dramatic decrease in incidence with HAART. Primary central nervous system lymphoma in HIV-infected patients represents a subset of EBV-associated ABC/immunoblastic DLBCL. HIV-associated DLBCL may also show centroblastic morphology and when it does the lymphoma is more likely to be of the germinal center subtype. Although EBV plays a role in some HIV-related centroblastic DLBCLs, this is much less frequent (30% of cases) than with the ABC/immunoblastic type.2
HIV-related Burkitt lymphoma (BL) is associated with immunodeficiency, but frequently develops earlier in the course of HIV infection and at higher levels of CD4 lymphocyte counts. As in non–HIV-infected patients, this lymphoma commonly involves extranodal sites. In most cases, the morphology and immunophenotype are similar to non–HIV-related BL, and isolated translocation of the MYC gene with an immunoglobulin gene is present by definition; however, morphologically atypical (eg, plasmacytoid) variants of BL may be seen in HIV-infected patients.2
This highly aggressive lymphoma was first recognized in HIV-infected patients and frequently seen in the oral cavity, but involvement of other sites in the gastrointestinal tract and in lymph nodes, lung, and skin may also be seen. Plasmablastic lymphoma exhibits features of mitotically active plasmablasts, including large centrally placed nucleoli; expression of CD79a, CD138, and MUM1; and absence of most B-cell–associated antigens. Cases involving the oral cavity are nearly all EBV associated, with significant but less common EBV positivity (∼75%) in other sites. MYC translocation is seen in approximately half of HIV-associated plasmablastic lymphomas.29,30
Primary Effusion Lymphoma
Primary effusion lymphoma (PEL) is another lymphoproliferative disorder first identified in HIV-infected patients.31,32 First called body cavity–based lymphoma, HIV-associated PEL was shown to exhibit several unique features, including exclusive involvement of body cavities in most patients, immunoblastic morphology, lack of B-cell–associated antigen expression, high rate of EBV positivity, and strong association with Kaposi sarcoma.33 A new herpesvirus first found in Kaposi sarcoma and called Kaposi sarcoma–associated herpesvirus or human herpesvirus 8 was also found to be present in essentially all cases of PEL.34,35 Over the years, PEL has been found to also involve non–body cavity sites, and a PEL-like neoplasm without any body cavity involvement was described and called extracavitary PEL. Although rare even in HIV-infected patients (<5% of HIV-related lymphomas), its incidence has not significantly decreased with HAART and its prognosis remains poor.36
Multicentric Castleman Disease and Other Related HHV8+ Lymphoproliferative Disorders
Multicentric Castleman disease (MCD) is a rare lymphoproliferative disorder with a strong association with HIV infection. This disorder involves polyclonal overproliferation of B lymphocytes and B-lymphocyte structures mostly in lymph nodes, based on overproduction of interleukin 6 (IL-6). These changes in HIV-infected patients are induced by poorly controlled infection of naïve plasmablastic B lymphocytes by HHV8, resulting in IL-6 overproduction, which in turn induces chronic overproliferation of plasma cells.37,38 The IL-6 overproduction in this setting is secondary to a viral IL-6 homolog gene introduced by HHV8 that drives production of viral IL-6 by the host cells. Although not a true neoplasm, HIV-associated MCD is characterized by severe constitutional symptoms and multisystem complications and typically requires treatment. Patients with HIV-related MCD are at risk of developing HHV8+ DLBCL with plasmablastic features derived from the HHV8-infected plasmablasts in MCD.38,39 Patients with MCD may also develop another unusual lymphoproliferative disorder, HHV8+ germinotropic lymphoproliferative disorder, an overproliferation of polyclonal HHV8+, EBV+ plasmablasts, which is more commonly seen in immunocompetent non–HIV-infected patients but may rarely be seen in HIV-infected patients.26 The incidence of HIV-associated MCD and HHV8+ DLBCL appears to be increasing in spite of HAART, although successful treatment of HIV-associated MCD may decrease the secondary development of HHV8+ DLBCL.37,39,40
Low-Grade Lymphoproliferative Disorders
Although most HIV-related lymphoproliferative disorders are high-grade, it has been increasingly recognized that certain low-grade B-cell disorders may also be related to HIV infection. EBV appears to play a major role in these low-grade B-cell conditions and although they are more common in other forms of immunodeficiency, such as the posttransplant setting, some are secondary to HIV. One example is EBV+ marginal zone lymphoma, usually of extranodal mucosa-associated lymphoid tissue (MALT) type but rarely other marginal zone lymphoma types.25,41,42 Another example is polymorphic B-cell lymphoproliferative disease, similar to posttransplant lymphoproliferative disorder.43,44
T-Cell Lymphoproliferative Disorders
Classic Hodgkin Lymphoma
The incidence of CHL in HIV-infected patients has always been higher than in the general population, but the association with HIV has not been strong enough to qualify CHL as an AIDS-defining neoplasm in these patients. HIV-related CHL exhibits important differences compared to non–HIV-related cases, including a much higher percentage of EBV positivity, more common mixed cellularity and lymphocyte-depleted subtypes, higher stage at presentation, and greater likelihood of extranodal involvement.27,48,49 Unlike many other HIV-associated lymphoproliferative disorders, the incidence of HIV-related CHL has significantly increased in the HAART era, from a relative risk of 10 pre HAART to a relative risk of 21 to 28 post HAART.48 HIV-associated CHL before HAART tended to develop earlier and with higher CD4 lymphocyte counts than most other HIV-related lymphomas. This supports the current thinking that development of CHL in HIV-infected patients treated with HAART is enhanced by the higher CD4 lymphocyte counts maintained in these patients.27 Before HAART, the prognosis of HIV-associated CHL was poor; however, in the HAART era the treatment and prognosis of CHL in these patients are similar to non–HIV-associated CHL.27,48
RESPIRATORY VIRAL PANDEMICS
Hematologic Impact of Respiratory Viral Pandemics
Although respiratory viral infections primarily affect the respiratory system, through direct infection of pulmonary epithelial cells by the virus and secondary bacterial infection involving respiratory sites, these viruses also impact other organs and systems. The respiratory effects are usually the major cause of morbidity and mortality; however, the nonrespiratory complications can also cause significant harm in these patients. Hematologic complications are among the most common of these nonrespiratory effects. Respiratory viral pandemics, including H1N1 influenza, SARS, MERS, and COVID-19, have all been associated with hematologic complications (Table 3). Some involve transient increases or decreases in circulating white blood cells, but the most impactful affect the coagulation system. Coagulopathy-related changes were some of the earliest reported nonrespiratory complications with COVID-19.4
Blood Cell Abnormalities in Respiratory Viral Pandemics
Lymphocytopenia is the most common blood cell abnormality in respiratory viral pandemics. This abnormality is seen in H1N1 influenza, SARS, MERS, and COVID-19 and may serve as a prognostic marker in some of these infections.50–53 In some (eg, COVID-19), the remaining circulating lymphocytes may exhibit reactive lymphoplasmacytic morphology, whereas, in others (eg, SARS) the lymphocytes lack reactive changes.6,53 The mechanism of lymphocytopenia in SARS and COVID-19 is thought to be related to direct infection of lymphocytes by the coronavirus agents using angiotensin-converting enzyme 2 (ACE2) receptors on the surface of lymphocytes and inducing a cytopathic effect.54 In addition, the inflammatory process and cytokine “storm” seen in these patients may induce lymphocyte apoptosis and phagocytosis by aberrantly activated macrophages.52 A decrease in CD8+ and other regulatory T lymphocytes, leaving a relative overabundance of proinflammatory T helper type 1 (Th1) CD4+ lymphocytes, is thought to contribute to the hyperinflammatory state in these patients.52 Natural killer (NK) cells are also known to be decreased in number and to exhibit an “exhausted” phenotype, probably induced by overproduction of certain cytokines and together resulting in decreased clearance of virus.52 Circulating neutrophils are increased in viral pandemic patients, and the absolute neutrophil count and neutrophil to lymphocyte ratio appear to have prognostic predictive significance.52 An increase in neutrophil extracellular traps is thought to produce tissue damage and help induce thrombosis.5,52 Thrombocytopenia is also observed in patients during these viral pandemics, and the cause is thought to be related to abnormal platelet production more so than severe disseminated intravascular coagulation (DIC) or other widespread platelet consumption.54 The decrease in production may be the result of abnormal thrombopoiesis related to direct infection of hematopoietic stem cells and/or bone marrow stromal cells by the respiratory viruses.54,55 Decreased production of platelets by pulmonary megakaryocytes secondary to lung tissue damage may also play a role.54,55 Morphologic and functional abnormalities in neutrophils and platelets are also seen and thought, in patients with COVID-19, to be the result of severe, transient abnormal hematopoiesis affecting these lineages.56
Coagulation Abnormalities in Respiratory Viral Pandemics
The clinically most significant nonrespiratory complications in respiratory viral pandemics are related to abnormal hemostasis, particularly those causing hypercoagulability and associated thrombotic events and related conditions.5,6,54 Patients during the H1N1, SARS, and COVID-19 pandemics have exhibited a significant hypercoagulable tendency and all have shown an increase in venous thromboembolism (VTE), as well as a smaller increase in arterial thrombosis.5 In addition to large vessel thrombosis, microthrombi may form in many organs and tissues, particularly in lung but also in other areas of the body. The cause of this hypercoagulable state is multifaceted. In patients with COVID-19 and possibly H1N1 influenza, microthrombus formation appears to be at least partially related to direct endothelial and microvascular damage caused by the virus directly infecting endothelial cells, which also express ACE2 receptors to allow viral entry, as well as endothelial inflammation.4,5,57 Endothelial cell infection and inflammation contribute to dysregulation of the kallikrein/kinin system and release and increased levels of coagulation factor VIII and vWF, along with local activation of platelets.5,54 Widespread microthrombi and microangiopathy are significantly more common in COVID-19 than in some other respiratory pandemics (H1N1) and may contribute to more severe pulmonary and other organ damage.5 The hyperimmune state and cytokine “storm” associated with these respiratory viruses also contribute to activation of the coagulation system. Elevated levels of fibrinogen, tissue factor, and other procoagulants and the presence of anti-phospholipid antibodies associated with the hyperimmune response are seen with H1N1 and COVID-19 infection.5 Although DIC may occur in some respiratory viral pandemic patients, particularly in those with secondary bacterial infections, severe DIC is not often seen as a direct complication of the virus. A significant elevation of D-dimer, independent of DIC, is seen in many of these patients and serves as a prognostic marker.5
The thrombotic complications associated with these infections may lead to significant morbidity and mortality. An increase in VTE and pulmonary embolus (PE) is seen in patients with N1H1 influenza, SARS, and COVID-19. In patients with COVID-19, VTE is reported in up to 69% and PE in up to 35% of seriously ill patients.5 Arterial thrombotic events are much less common, but still increased in incidence and may have a devastating impact on the patient. With COVID-19, arterial thrombosis occurs in approximately 4% of critically ill patients, exhibits a wide distribution of sites (limb arteries in 39%, cerebral arteries in 24%, great vessels in 19%, coronary arteries in 9%, and superior mesenteric artery in 8%), and is associated with a mortality rate of 20%.57 The recognition of frequent thrombotic complications with the COVID-19 pandemic has led to an increased use of thromboprophylaxis for these patients, but the degree of improvement in the incidence of these complications is still uncertain.5,6
As a result of increased crowding in parts of the world, proximity to nonhuman viral disease vectors, and the ease of international travel, viral pandemics have increased in incidence during the past few decades. Each of these widespread infections has been associated with multiple hematologic complications, and these complications are often responsible for significant morbidity and mortality in infected patients. In the case of the HIV pandemic, introduction of HAART brought on a decrease in the incidence of many HIV-related hematologic disorders; however, essentially all of them continue to be more common than in the general population and some have actually increased in incidence following HAART. Respiratory viral pandemics (H1N1 influenza, SARS, MERS, and COVID-19) have been associated with blood cell and coagulation abnormalities, both contributing significantly to the clinical course and outcome in affected patients. Improved understanding of the pathogenesis of the various hematologic complications associated with viral pandemics has led to prevention and/or effective treatment of some of them, but more work needs to be done to fully protect patients from the hematologic impact of these infections. The experience and knowledge gained in recent years will likely help lessen the impact of hematologic complications in inevitable future viral pandemics.
The author has no relevant financial interest in the products or companies described in this article.
Presented in part at the Seventh Princeton Integrated Pathology Symposium; May 16, 2020; virtual.