Postsplenectomy Cytomegalovirus Mononucleosis is a Distinct Clinicopathologic Syndrome

CASE REPORT

Postsplenectomy Cytomegalovirus Mononucleosis is a Distinct Clinicopathologic Syndrome Xiang Y. Han, MD, PhD, Beth A. Hellerstedt, MD and Charles A. Koller, MD

Abstract: Lymphocytosis in response to viral infection, such as infectious mononucleosis, rarely exceeds 20 ⫻ 109/L in the adult population. Transfusion-acquired cytomegalovirus (CMV) mononucleosis after trauma-related splenectomy may cause prominent lymphocytosis, but the history and timing usually hint at the diagnosis. We describe a case of severe CMV mononucleosis that was acquired naturally decades after splenectomy. Together with the 2 similar cases that we reported recently, these cases all presented as initial diagnostic challenge because of a remote history of splenectomy, a prolonged febrile illness (⬃4 weeks), marked lymphocytosis (peak 27.9 ⫻ 109/L), and undetectable or weakened anti-CMV IgM antibody response. The diagnosis was eventually established through detection of circulating CMV antigen or DNA and a year or longer follow-up with serial determination of IgM and IgG antibodies. Two similar cases were also identified in the literature and reviewed. Although the impaired IgM response may confuse the diagnosis, it correlates well with recent studies showing that human blood IgM memory B cells are circulating splenic marginal zone B cells; asplenic or splenectomized individuals, irrespective of the underlying cause, have undetectable IgM memory B cells. Together, these findings suggest that distant or recent postsplenectomy CMV mononucleosis represents a distinct clinicopathologic syndrome resulting from poor control of early viremia because of the lack of both splenic filtration and the typical brisk IgM response. For the practicing clinician, recognizing these features may aid timely diagnosis and treatment. Key Indexing Terms: Cytomegalovirus; Infectious mononucleosis; Splenectomy; IgM response; Marked lymphocytosis. [Am J Med Sci 2010;339(4):395–399.]

ytomegalovirus (CMV) is a ubiquitous large ␤-herpesvirus and causes a variety of primary and secondary infections, such as congenital neonatal infections, infectious mononucleosis in healthy individuals, and reactivation in immunocompromised patients.1 Primary CMV infection manifested as mononucleosis occurs mainly in school age children and is usually subclinical. A recent serosurvey shows that 61% to 73% of German adults, ranged from 20 to more than 60 years old, have detectable CMV antibodies.2 The same study and one from the United States3 have also shown that, in some adult patients (18 –56 years of age), CMV mononucleosis may present with a prolonged clinical course. CMV is responsible for most of the Epstein Barr virus (EBV)-negative cases of infectious mononucleosis. CMV mononucleosis after splenectomy has been reported. It typically occurs within 2 to 4 weeks after trauma

C

From the Department of Laboratory Medicine (XYH), The University of Texas M. D. Anderson Cancer Center, Houston, Texas; Texas Oncology PA (BAH), Austin, Texas; and Department of Leukemia (CAK), The University of Texas M. D. Anderson Cancer Center, Houston, Texas. Submitted November 12, 2009; accepted in revised form December 9, 2009. Correspondence: X. Y. Han, MD, PhD, Department of Laboratory Medicine, Unit 84, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030 (E-mail: [email protected]).

surgery and multiple blood transfusions, and a marked lymphocytosis is a feature.4 – 6 However, the practice of using leukocyte-reduced blood transfusions in the past 2 decades has made this disease rare.7 Recently, we reported 2 cases of severe CMV mononucleosis that initially presented as a diagnostic dilemma because of a remote history of splenectomy, a prolonged febrile illness, and marked lymphocytosis (peak 27.9 ⫻ 109/L).8 Both patients were referred to rule out chronic lymphocytic leukemia, and bone marrow examination also noted T-cell receptor (TCR)-␥ gene rearrangements, suggesting monoclonal T-cell proliferation. Because the acute phase had passed, the CMV pp65 antigenemia test, routinely performed in our institution, was negative. Eventually, the diagnosis of CMV mononucleosis was established by demonstrating CMV DNA with polymerase chain reaction (PCR) in a retrieved acute-phase blood DNA sample in 1 case and serial determination of antiCMV antibodies and exclusion of other causes of infectious mononucleosis, such as EBV, Toxoplasma, and other viruses in both cases. Surprisingly, close follow-up also revealed lack of anti-CMV IgM response in 1 case and weakened and delayed response in another, which contrasted with an augmented IgG response in both cases. The poor IgM response was paradoxical to the prominent and prolonged lymphocytosis. In this study, we extend our experience by describing a similar case of postsplenectomy CMV mononucleosis and reviewing 2 additional cases in the literature. The clinical features of the 5 cases correlate with relevant experimental studies, which led us to the impression that the postsplenectomy CMV mononucleosis represents a distinct clinicopathologic syndrome.

CASE REPORT The patient was a 34-year-old man who had a medical history of splenectomy at teenage because of idiopathic thrombocytopenic purpura. Since then, he had been in good health, and he had visited his physician regularly. He developed a sudden onset of fever up to 39.5°C, headache, myalgias, arthralgias, and night sweats; after failing to respond to 2 courses of oral antibiotics, he was hospitalized elsewhere for the illness. Admission physical examination documented the fever and disclosed a grade 2/6 holosystolic murmur heard loudest at the apex. Blood cultures were negative, and a transesophageal echocardiogram showed mitral valve prolapse and regurgitation without any evidence of endocarditis. A computed tomographic scan revealed several cervical lymph nodes, up to 14 mm in size, and a possible accessory spleen at the prior surgical site. A prominent leukocytosis with marked absolute lymphocytosis was observed (Figure 1), which led to a flow cytometric analysis of the peripheral blood 14 days after onset of illness, revealing proliferation of large granular lymphocytes (LGL) with an unusual T-cell phenotype, because of the loss of CD5 and CD7 markers, and increased natural killer (NK) cells with expression of CD56 and CD16. The CD4 to CD8 ratio was inverted. Because the white blood cell count (WBC) continued to rise and reached a peak of 33.8 ⫻ 109/L with an absolute

The American Journal of the Medical Sciences • Volume 339, Number 4, April 2010

395

Han et al

Cell count, x1000/ul

25 Lymphocytes

20

Neutrophils Monocytes

15 10 5 0 0

7

14

21 28

35 42

49

56 63

70 77

84 91

98 105

Days after onset

FIGURE 1. Prominent lymphocytosis in CMV mononucleosis after remote splenectomy.

lymphocyte count of 24.9 ⫻ 109/L (day 18 after onset; Figure 1), the patient was referred to our institution with a diagnosis of possible T-cell acute lymphoblastic leukemia. When seen at our institution, initial laboratory examination (day 21 after onset) revealed a WBC of 27.5 ⫻ 109/L, including 19.0 ⫻ 109 lymphocytes/L and 2.2 ⫻ 109 monocytes/L, and elevated transaminases [alanine transaminase, 270 IU/L (normal range 7–56 IU/L); aspartate transaminase, 155 IU/L (normal range, 15– 46 IU/L)] and lactate dehydrogenase [1505 IU/L (normal range, 313– 618 IU/L)]. A bone marrow biopsy and aspirate showed trilineage hematopoiesis with an increase in CD3⫹, CD8⫹ cytotoxic T-cells/LGL (⬃68% of lymphocytes), with 50% of the CD8⫹ cells showing expression of TCR V-␤ 5.1. TCR ␤ gene rearrangements were also demonstrated by PCR, suggesting monoclonal or oligoclonal T-cell proliferation. Together, these findings indicated a possible T-cell lymphoproliferative process. Viral etiologies were also sought to explain the clinical and laboratory findings. Although the tests for hepatitis B and C viruses, HIV, and HTLV I/II were negative, and the antiEBV antibodies represented past infection, a CMV pp65 antigenemia test (21 days after onset of illness) was positive for 2 cells/106 WBC, consistent with acute CMV infection in the clinical context. With improvement of the constitutional symptoms 25 days after onset, the patient was followed up closely without anti-CMV treatment, and his lymphocytosis declined remarkably in the following month (Figure 1). A follow-up CMV antigenemia test (53 days after onset) was negative. However, the total anti-CMV antibodies by the latex agglutination method (BD Microbiology Systems, Sparks, MD) were negative repeatedly at 25 and 53 days after onset. By day 62, an enzyme immunoassay for the anti-CMV IgM and IgG antibodies (Diamedix, Miami, FL) was weakly positive for IgM (index 1.96, cutoff 0.90) but strong for IgG (index 7.90, cutoff 0.90). These results, therefore, confirmed CMV mononucleosis. At 1-year follow-up, the patient’s lymphocyte count was 5.6 ⫻ 109/L, with essentially normal blood chemistry. The anti-CMV IgM turned negative, but IgG remained strong (index 8.79) along with affinity maturation. Throughout the course, the patient’s neutrophil counts remained normal, but monocyte counts were moderately elevated (peak 3.7 ⫻ 109/L; Figure 1). Literature Review and Discussion A primary function of the spleen is to filter senescent red cells and particulate substances in the bloodstream. As such, it is a major site for primary immunological response to bloodborne antigens. CMV is a blood-borne pathogen as indicated by

396

the presence of viremia during its primary infection and reactivation. For eusplenic patients with CMV mononucleosis, leukocytosis occurs infrequently (4%–14% of cases), and the absolute lymphocyte count remains normal to mildly elevated (⬍8 ⫻ 109/L) despite common relative lymphocytosis.2,3 Splenomegaly can be detected by ultrasound examination in 83% of cases.2 This feature, along with reported cases of spontaneous splenic rupture with identification of CMV inclusions in the spleen,9 suggests that the spleen is a replication site of and, hence, early immunity against CMV mononucleosis. Furthermore, primary CMV infection elicits early strong IgM response,10 being positive diagnostically in more than 99% of cases and strongly positive in 62% to 64%.2,3 Typically, the anti-CMV IgM rises from being weakly positive several days after onset to strongly positive in a few weeks.3,11 It is well known that asplenic or splenectomized individuals are prone to life-threatening infections by encapsulated bacteria. They also mount poor antibody response to bacterial polysaccharide vaccines, and study shows that it is the IgM response that is impaired, not IgG.12 Recent data further suggest that this is because of the lack of IgM memory B cells.13 Two additional studies corroborate such IgM defect after splenectomy by showing that human blood IgM memory B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire14 and that asplenic or splenectomized individuals, regardless of the underlying causes, have undetectable IgM memory B cells.15 Our case was worrisome because of prolonged fever and marked lymphocytosis (mostly LGL and NK cells). It was difficult to connect this picture clinically with his remote history of splenectomy. Historically, multiple blood transfusions after trauma-related splenectomy may cause severe CMV mononucleosis within 2 to 4 weeks with an unusually high lymphocytosis.4 – 6 However, this disease is now much less common with the practice of leukocyte-reduced red cell transfusion. Besides, the immediate history of trauma surgery and transfusion usually render the diagnostic clues. Our patient’s surgery was remote, and he had received no transfusions. Although it is common to see increases of LGL and NK cells in usual CMV mononucleosis,16 typical levels are far below the numbers seen in our case. In addition, our patient’s LGL were also aberrant in view of the loss of CD5 and CD7 expressions. More perplexingly, TCR gene rearrangements were also found, hinting clonal T-cell proliferation. Fortunately, the CMV pp65 antigenemia test was most helpful in reaching a presumptive diagnosis on referral to our institution. We routinely use antigenemia to monitor CMV reactivation in our stem cell recipients and in patients with leukemia and lymphoma.17,18 This test has a turnaround time of several hours and is highly specific. Over the years, we have also found it useful to diagnose primary CMV infection.11 With the current case included, there are at least 9 reported cases of primary CMV infection diagnosed by antigenemia test, and positive cells can be detected from 3 to 32 days after onset of illness.11,19,20 The presence of 2 cells/106 WBC at 21 days after onset in our case was probably at the tail of antigenemia (viremia), following the peak of lymphocytosis at 18 days after onset. Earlier tests may have detected more positive cells.11 The anti-CMV IgM response in our patient was delayed and weakened on repeated testing. This feature is paradoxical to the marked and year-long proliferation of T cells and B cells and the strong IgG antibody response. As discussed above, this defect is well explained by the loss of splenic IgM memory B cells as a consequence of splenectomy. Without adequate IgM, Volume 339, Number 4, April 2010

Postsplenectomy CMV Mononucleosis

TABLE 1. Clinical and pathologic features of 5 cases of postsplenectomy CMV mononucleosis Feature General and history Age (yr), sex Referral diagnosis Chief complaint Splenectomy for and when Residual spleen Transfusion before onset Significant laboratory tests Peak lymphocytes (⫻109/L) Peak monocytes (⫻109/L) ALT (IU/L) AST (IU/L) LDH (IU/L) Key CMV tests CMV viremia by IgM peak index and time IgG peak index and time Other antibody tests Epstein-barr virus Hepatitis B and C viruses Toxoplasma HIV Adenovirus Blood flow cytometry Bone marrow study Additional findings Final diagnosis Anti-CMV treatment Outcome Source of infection

This case

Case 28

Case 38

Case 420

Case 53

34, M NK cell lymphocytosis Fever ⫻ 3 wk

37, M Marked lymphocytosis Fever ⫻ 3 wk

54, M CLL

36, M Not applicable

43, M Not applicable

Fever ⫻ ⬃4 wk

Fever, headache

ITP, teenage

HS, age 4 years

HS, childhood

Likely on CT scan

Unknown

No

Likely, lifelong bilirubinemia No

Fever, visual change Trauma, 2 d earlier Unknown

No

2 Packs of filtered RBC

No

24.9, d 18

27.9, ⬃d 14

25.1, ⬃d 14

24.5, ?d

Unknown

3.7, d 18

3.74, ⬃d 14

1.55, ⬃d 14

Unknown

Unknown

326, d 18; 270, d 21 195, d 18 531, d 18; 1505, d 21

146, d 21

Unknown

362

Unknown

Unknown 1215, d 21

Unknown Unknown

218 441

Unknown Unknown

Antigenemia, d 21

PCR, ⬃d 14

Weak and late (see text) Strong (see text)

Weak and late (3.0 at wk 14) Strong (9.0 at wk 3)

Negative at referral (9 wk) Negative (wk 10 and 50) Strongest (35.6 at wk 10)

Antigenemia and PCR Negative 2 wk after onset Strong 2 wk after onset

Past infection Negative

Past infection Negative

Past infection Negative

Negative Negative

Past infection Likely ND

ND Negative ND Up NK (CD56⫹ CD16⫹) Up CD8⫹

Negative Negative ND Up CD3⫹, CD8⫹, CD57⫹ Up NK (CD56⫹ CD16⫹/CD3⫺) TCR␥ gene rearranged CMV mononucleosis No Resolved at 1 year ? From young children

Negative Negative Negative Up CD3⫹, CD8⫹, CD57⫹, D56⫹ Normal at wk 9

Negative Negative ND Up CD8⫹

Likely ND Pre-HIV era Likely ND Likely ND

Reactive hypercellularity Retinitis, severe hepatitis CMV mononucleosis Ganciclovir Resolved in 1 mo Incubation of CMV at surgery

Unknown

TCR␤ gene rearranged CMV mononucleosis No Resolved at 1 year ? From child contact

TCR␥ gene rearranged CMV mononucleosis No Resolved at 1 year ?

Trauma, age 17 years Unknown

Likely ND Weak and late (see text) Strong (see text)

Likely ND CMV mononucleosis Unknown Resolved at 1 year Unknown

ALT, alanine transaminase; AST, aspartate transaminase; CLL, chronic lymphocytic leukemia; CT, computed tomography; HS, hereditary spherocytosis; ITP, idiopathic thrombocytopenic purpura; LDH, lactate dehydrogenase; ?, uncertain or unknown; ND, not done; PCR, polymerase chain reaction; RBC, red blood cells; TCR, T-cell receptor; Up means elevated cluster designated (CD) cells by flow cytometry.

CMV causes prolonged illness, whereas the host attempts to compensate by an overwhelming cytotoxic T-cell response and eventually a strong IgG component. The current case is very similar to the 2 cases we recently reported,8 as summarized in the Introduction and Table © 2010 Lippincott Williams & Wilkins

1. In all 3 cases, the natural history of the disease is demonstrated because the diagnosis was much delayed and antiviral treatment was not given. The constitutional symptoms and signs, such as fever, headache, malaise, and hepatitis resolved about 4 weeks after onset; marked lymphocytosis (⬎10 ⫻

397

Han et al

109/L) improved by 6 to 8 weeks in a rapid rise and fall fashion; but normalization of absolute lymphocytosis (⬍4.8 ⫻ 109/L) took about a year. The TCR ␤ or TCR ␥ gene rearrangements in our cases are noteworthy. Compared with the sustained clonal TCR gene rearrangements in lymphoid malignancies,21 such changes in our cases are transient and limited to T-cell subsets. Similar findings have been reported in mononucleosis caused by CMV22,23 and EBV.24 In addition, recent experimental data show that TCR ␤ chain shares in human CD8⫹ T-cell responses to CMV and EBV and that CMV infection induces a rapid change in the expression of NK cell receptors on CD8⫹ T cells.25,26 Therefore, these findings are consistent with the concept that cytotoxic T cells ultimately control CMV. In an earlier report of 82 cases of CMV mononucleosis, a case after distant splenectomy was briefly described.2 This 43-year-old man, with a history of splenectomy for trauma at the age of 17 years, also manifested prolonged fever and impaired IgM response, according to our analysis of the data. This patient’s serial anti-CMV IgM titers (by an immunofluorescence assay) were 1:32, 1:64, 1:128, and 1:⬍16, respectively at 16, 42, 89, and 402 days after onset, but the corresponding IgG titers (by complement fixation) were 1:64, 1:2048, 1:1024, and 1:256. In contrast, for all 18 eusplenic cases (age 16 –56 years, mean 30 years) with serial antibody tests, the median IgM and IgG titers were both 1:256, and they were at 18 and 32 days after onset, respectively. Thus, this patient’s IgM response, rising slowly to 1:128 at 89 days— half of the median of control cases and only 1 of 16 of his own IgG peak at 42 days—was weak and much delayed. Also, compared with all 81 cases with positive IgM, this patient’s titer was below the 8th percentile at 16 days or below the 13th percentile at its peak. However, his IgG peak was the second highest of all, rising by 32-fold in a few weeks. Therefore, this antibody response pattern, determined by different methods, is identical to that of our 3 cases.3 Recently, Assy et al20 reported a case of a 36-year-old man who developed persistent fever 2 days after splenectomy for blunt trauma and transfusion of 2 packs of leukocytereduced red cells. The patient manifested marked lymphocytosis (peak 24,500/␮L), severe hepatitis on liver biopsy, and retinitis. The diagnosis of CMV mononucleosis was established by positive antigenemia (peak 125 cells/106 WBC) and PCR tests, which led to timely treatment with ganciclovir and recovery a month after onset. In search of the source of infection, the removed spleen was found to contain CMV, suggesting incubation of the virus just before the surgery and transfusion. Therefore, in view of the high level of antigenemia and retinitis, the latter being hardly seen in usual mononucleosis, this patient’s disease was nearly fulminant and brought about by splenectomy. Notwithstanding the lymphocytosis and severity, the patient had no detectable anti-CMV IgM in contrast to a strong IgG response at 2 weeks after onset. This result did confuse the initial diagnosis of primary CMV disease versus reactivation, as reflected in the title of the case report, because the latter is characterized by negative IgM but positive IgG because of past exposure. However, CMV reactivation is associated with lymphopenia and not lymphocytosis, a remarkable distinction. After primary exposure, normal immunity against CMV is robust; a recent study finds that the human immune system devotes up to 10% of the memory repertoire of CD4⫹ and CD8⫹ T cells to curb the virus from reactivation.27 The clinical and pathologic features of these 5 cases are summarized in Table 1. In addition, in the study by Baumgart-

398

ner et al5 on transfusion-acquired CMV disease after splenectomy, 2 of the 3 patients who were tested for anti-CMV IgM were also negative for this response. Together, these cases suggest that CMV mononucleosis after splenectomy, either distant or recent, is a unique clinicopathologic syndrome with prolonged fever (⬃4 weeks), moderately severe hepatitis, marked lymphocytosis (⬎10 ⫻ 109/L) for 6 to 8 weeks with a peak of ⬃28 ⫻ 109/L, and impaired (no or low and delayed) IgM response. The presence of accessory or residual spleen may explain the low-level IgM in some cases, rather than no IgM response at all. These cases unravel an important aspect of the humoral immunity against CMV, ie, the crucial roles of the spleen in producing IgM and in coordinating B-cell and T-cell responses. Severe viral infection after splenectomy is rarely known or reported. Thus, the earlier report by Baumgartner et al5 and our delineation of this disease entity represent, to our knowledge, the only descriptions of defective IgM response during actual infection by a microorganism (virus or bacterium). Therefore, the spleen fends off blood-borne pathogens, such as encapsulated bacteria and CMV, through both innate immunity (filtration and phagocytosis) and acquired immunity (antibody production). As reviewed by Racine and Winslow,28 extensive recent experimental data derived from knockout animals also demonstrate that IgM plays an essential yet underappreciated role in the defense against microbial infections. Because it is common to diagnose acute CMV mononucleosis by detecting an early and brisk IgM response, lack of such response after splenectomy may lead to misinterpretation of test results and potentially delayed or wrong diagnosis. Thus, knowing this defect of “splenectomy—no or low IgM” may prompt tests to detect circulating CMV through antigenemia, PCR, or culture. Culturing CMV by the shell vial method carries a turnaround time of ⬃20 hours, but the sensitivity is slightly lower.1,17 In the meantime, the triad presentation of a history of remote or recent splenectomy, prolonged fever with hepatitis, and rapidly rising lymphocytosis (10 –30 ⫻ 109/L) should alert the clinician to the possibility of CMV mononucleosis. Indeed, it might have been possible for our patient to be treated for acute leukemia with disastrous consequences. On the other hand, early anti-CMV therapy can be considered to hasten recovery in future case of such infection. REFERENCES 1. Pass RF. Cytomegalovirus. In: Knipe DM, Howley PM, Griffin DE, et al, editors. Fields virology, 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2001. p. 2675–705. 2. Just-Nubling G, Korn S, Ludwig B, et al. Primary cytomegalovirus infection in an outpatient setting—laboratory markers and clinical aspects. Infection 2003;31:318 –23. 3. Horwitz CA, Henle W, Henle G, et al. Clinical and laboratory evaluation of cytomegalovirus-induced mononucleosis in previous healthy individuals: report of 82 cases. Medicine (Baltimore) 1986;65: 124 –34. 4. Okun DB, Tanaka KR. Profound leukemoid reaction in cytomegalovirus mononucleosis. JAMA 1978;240:1888 –9. 5. Baumgartner JD, Glauser MP, Burgo-Black AL, et al. Severe cytomegalovirus infection in multiply transfused, splenectomized, trauma patients. Lancet 1982;2:63– 6. 6. Langenhuijsen MMAC, van Toorn DW. Splenectomy and the severity of cytomegalovirus infection. Lancet 1982;2:820. 7. Blajchman MA. The clinical benefits of the leukoreduction of blood products. J Trauma 2006;60(6 suppl):S83–90. 8. Han XY, Lin P, Amin HM, et al. Post-splenectomy cytomegaloviral

Volume 339, Number 4, April 2010

Postsplenectomy CMV Mononucleosis

mononucleosis: marked lymphocytosis, TCR-gamma gene rearrangements, and impaired IgM response. Am J Clin Pathol 2005;123:612–7.

18. Han XY. Epidemiologic analysis of reactivated cytomegalovirus antigenemia in patients with cancer. J Clin Microbiol 2007;45:1126 –32.

9. Alliot C, Beets C, Besson M, et al. Spontaneous splenic rupture associated with CMV infection: report of a case and review. Scan J Infect Dis 2001;33:875–7.

19. Lesprit P, Scieux C, Lemann M, et al. Use of cytomegalovirus (CMV) antigenemia assay for the rapid diagnosis of primary CMV infection in hospitalized adults. Clin Infect Dis 1998;26:646 –50.

10. Kangro HO, Griffiths PD, Huber TJ, et al. Specific IgM class antibody production following infection with cytomegalovirus. J Med Virol 1982;10:203–12.

20. Assy N, Gefen H, Schlesinger S, et al. Reactivation versus primary CMV infection after splenectomy in immunocompetent patients. Dig Dis Sci 2007;52:3477– 80.

11. Han XY, Anderlini P, Tarrand J. Diagnosis and monitoring of primary cytomegalovirus infection by antigenemia test in stem cell donor. Bone Marrow Transplant 2002;30:411.

21. Vega F, Medeiros LJ, Jones D, et al. A novel four-color PCR assay to assess T-cell receptor gamma gene rearrangements in lymphoproliferative lesions. Am J Clin Pathol 2001;116:17–24.

12. Molrine DC, Siber GR, Samra Y, et al. Normal IgG and impaired IgM responses to polysaccharide vaccines in asplenic patients. J Infect Dis 1999;179:513–7. 13. Carsetti R, Pantosti A, Quinti I. Impairment of the antipolysaccharide response in splenectomized patients is due to the lack of immunoglobulin M memory B cells. J Infect Dis 2006;193:1189 –90. 14. Weller S, Braun MC, Tan BK, et al. Human blood IgM “memory” B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire. Blood 2004;104:3647–54. 15. Kruetzmann S, Rosado MM, Weber H, et al. Human immunoglobulin M memory B cells controlling Streptococcus pneumoniae infection are generated in the spleen. J Exp Med 2003;197:939 – 45. 16. Kano Y, Shiohara T. Current understanding of cytomegalovirus infection in immunocompetent individuals. J Dermatol Sci 2000;22:196 – 204. 17. Nicholson VA, Whimbey E, Champlin R, et al. Comparison of cytomegalovirus antigenemia and shell vial culture in allogeneic marrow transplantation recipients receiving ganciclovir prophylaxis. Bone Marrow Transplant 1997;19:37– 41.

© 2010 Lippincott Williams & Wilkins

22. Mathew P, Hudnall SD, Elghetany MT, et al. T-gamma gene rearrangement and CMV mononucleosis. Am J Haematol 2001;66:64 – 6. 23. Delobel P, Godel A, Thebault S, et al. Transient clonal expansion of T-large granular lymphocytes during primary cytomegalovirus infection. J Infect 2006;53:e65–7. 24. Strickler JG, Movahed LA, Gajl-Peczalska KJ, et al. Oligoclonal T cell receptor gene rearrangements in blood lymphocytes of patients with acute Epstein-Barr virus-induced infectious mononucleosis. J Clin Invest 1990;86:1358 – 63. 25. van Stijn A, Rowshani AT, Yong SL, et al. Human cytomegalovirus infection induces a rapid and sustained change in the expression of NK cell receptors on CD8⫹ T cells. J Immunol 2008;180:4550 – 60. 26. Venturi V, Chin HY, Asher TE, et al. TCR beta-chain sharing in human CD8⫹ T cell responses to cytomegalovirus and EBV. J Immunol 2008;181:7853– 62. 27. Sylwester AW, Mitchell BL, Edgar JB, et al. Broadly targeted human cytomegalovirus-specific CD4⫹ and CD8⫹ T cells dominate the memory compartments of exposed subjects. J Exp Med 2005;202:673– 85. 28. Racine R, Winslow GM. IgM in microbial infections: taken for granted? Immunol Lett 2009;125:79 – 85.

399