Thrombocytopenia is an infrequent yet well-recognized complication of viral infections like mumps, rubella, varicella, Parvovirus, and infectious mononucleosis by Epstein-Barr virus.1 Although thrombocytopenia has been commonly observed in patients with congenital cytomegalovirus (CMV) infection, it has rarely been reported after CMV infection in immunocompetent children or adults.2-4 The most common form of CMV infection in the immunocompetent is asymptomatic and usually detected retrospectively. It is a well-recognized cause of illness characterized by myalgia, malaise, headache, fever, and sore throat. Peripheral blood examination shows the atypical mononuclear cells present in other infections, as Epstein-Barr virus infection, toxoplasmosis, or Q fever. The illness is usually self-limiting, and serious complications (such as hemolysis, Guillain-Barré syndrome, granulomatous hepatitis, carditis, pneumonia, meningoencephalitis, and thrombocytopenia) are rare. In the literature, approximately 20 cases of CMV-correlated thrombocytopenia are reported.5-12 Most of these patients were treated with corticosteroids, but some recent experiences have recommend the use of intravenous immunoglobulin, particularly in cases of severe bleeding or evidence of purpura and petechiae or when the platelet level is very low (<20 × 109/L).13,14 We report a case of an immunocompetent child with acquired CMV infection presenting with purpura and thrombocytopenia in the absence of other symptoms and treated with intravenous immunoglobulin.
A 7-year-old child presented with a 2-day history of purpura on his limbs and trunk, in the absence of other symptoms. Three weeks ago, he presented with fever (39°C) for 3 days, with a successive spontaneous regression, without use of medications. He had no history of illnesses that would cause an immunodeficient state, and he had no risk factors for infection with human immunodeficiency virus. On physical examination, he had a temperature of 36°C, an extensive purpura over his limbs and trunk and a spleen palpable 1 cm below the costal margin. There was no lymphadenopathy and hepatomegaly. Laboratory test revealed a leukocyte count of 8.9 × 109/L (8900 per μL), with 0.24 neutrophils, 0.63 lymphocytes (including many atypical mononuclear forms), 0.06 monocytes, 0.04 eosinophils, and 0.01 basophils. The hemoglobin level was 13.5 g/dL, and platelet count was 17 × 109/L (17,000 per μL). Prothrombin time, partial thromboplastin time, erythrocyte sedimentation rate, and reaction chain protein were normal. Blood chemistry values resulted in normal values, with the exception of lactate dehydrogenase (372 U/L [reference range, 100-260]), aspartate aminotransferase (80 U/L [reference range, 2-40), and alanine aminotransferase (211 U/L [reference range, 2-40]). A chest roentgenogram was unremarkable, and urine cultures were sterile. Results of serological tests for Brucella species, Salmonella typhi and Salmonella paratyphi, Epstein-Barr virus, and Toxoplasma were negative. Hepatitis screen, including hepatitis A, B, and C virus, autoantibodies including antinuclear antibody, rheumatoid factor, and anti-double-stranded DNA were negative. Sternal marrow appearances were consistent with thrombocytopenia because of peripheral consumption, whereas megakaryocyte morphology was normal. Immunoglobulin (Ig) M and IgG anti-CMV, determined by a commercial enzyme-linked immunosorbent assay test, resulted positive: IgM titer was 6.046 arbitrary units (UA)/mL (positive, >0.5), whereas IgG titer was 174 UA/mL (positive, >15). Serum CMV-DNA was determined by a real-time polymerase chain reaction (sensibility, >100 copies/mL), and it resulted positive (13,400 copies/mL). Treatment with intravenous γ-globulin of 1 g/kg per day was started shortly after admission and was prolonged for 2 days. The 2 infusion of high-dose intravenous immunoglobulin produced a rapid response; in fact, after 24 hours of treatment, the platelet count was 89×109/L, with a complete regression of purpura on limbs and trunk. On day 7, the count was 179 × 109/L. After hospitalization, the patient was examined monthly. After 1 month, the patient was asymptomatic, but we observed a new reduction in the number of platelets (41 × 109/L). In relation to this new reduction, serum CMV DNA was determined and resulted positive (12,900 copies/mL). Immunoglobulin M titer was 6.012 UA/mL. After 2 months, we observed a spontaneous increase in the number of platelets (98 × 109/L), in concomitance with an evident reduction of serum CMV DNA (2,300 copies/mL) and IgM titer (2.121 UA/mL). At the third month, the number of platelets was 110 × 109/L serum CMV DNA was positive, with a loss viral load (900 copies/mL), and IgM titer was 0.921 UA/mL. After 4 months, the platelet count returned to normal values (280 × 109/L), serum CMV DNA became undetectable, and IgM titer was 0.631 UA/mL. At the sixth month, the number of platelets became stabilized at 320 × 109/L, and IgM titer was negative (0.3 UA/mL). The statistical analysis has evidenced a significant correlation of platelet count with IgM titer (P = 0.007) and serum CMV DNA levels (P = 0.006). P < 0.05 indicated a significant correlation. Table 1 evidences the platelet count, IgM anti-CMV titer, and serum CMV DNA level at onset and during follow-up.
Although thrombocytopenia has been seen in patients infected by human immunodeficiency virus with concomitant CMV infection and has been commonly observed in patients with congenital CMV infection, CMV-induced thrombocytopenia in immunocompetent children and adults seems to be rare; in fact, in the literature, only approximately 20 previous cases have been reported.5-12 We report a case of an immunocompetent child with acquired CMV infection presenting with purpura and thrombocytopenia in the absence of other symptoms and treated with intravenous γ-globulin. Because CMV-induced thrombocytopenia has been reported so rarely in immunocompetent children, it is very difficult to know the mechanism by which CMV induces the reduction of platelet count and the natural course of this disease. About the pathogenesis, 2 possible mechanisms have been proposed: a direct cytopathic effect of CMV on megakaryocytes and an indirect immune-mediated effect.5 Actually, in the major part of previous reported cases, the indirect effect is considerably responsible, in relation to the temporal correlation between symptomatic thrombocytopenia and the onset of CMV infection. In fact, in many previous experiences, symptomatic thrombocytopenia was observed several weeks after the onset of infection.5,6 This temporal relationship seems to evidence that an indirect mechanism is responsible for thrombocytopenia; in fact, a direct effect of CMV would be expected to be evident much sooner in the clinical course. Also, this case report seems to support an immune-mediated mechanism: our child, 3 weeks before comparison of purpura, presented with fever for 3 days-probably an expression of the onset of CMV infection. The distinction between direct and indirect effect is fundamental because it may have very important therapeutic implications: an immune-mediated effect would be expected to respond to steroids (prednisone), but a cytopathic effect would not; the direct mechanism presupposes a good response to specific anti-CMV treatment with ganciclovir. However, the review of the literature evidences that corticosteroids are effective only in approximately one third of cases, with results ranging from no response to normal platelet counts after 2 weeks of treatment, and specific anti-CMV treatment with ganciclovir is of unproven efficacy and toxicity.9,14 Also, splenectomy, useful in the patients affected by idiopathic thrombocytopenic purpura, is relatively ineffective for CMV-induced thrombocytopenia.14 Some recent case reports have recommend the use of intravenous immunoglobulin in cases of severe bleeding or evidence of purpura and petechiae or when the platelet level is very low (<20 × 109/L). The use of high-dose intravenous γ-immunoglobulin for the treatment of immune thrombocytopenic purpura (ITP) was first reported more than 2 decades ago. After the therapeutic benefit of intravenous γ-immunoglobulin was established in ITP, it was then successfully used to treat many other autoimmune diseases. Although a complete definition of the mechanism of intravenous γ-immunoglobulin action is still lacking, extensive research suggests that intravenous γ-immunoglobulin may achieve its therapeutic effects through multiple mechanisms. Intravenous γ-immunoglobulin exerts immunomodulatory effects that may include anti-idiotypic neutralization of antiplatelet antibodies, stimulation of Fcγ receptor IIB expression, and inhibition of Fcγ receptor-mediated platelet destruction.15 Many studies evidence that patients affected with ITP and treated with intravenous γ-immunoglobulin, alone or in combination with corticosteroids, demonstrate a high percentage of response, with increasing rapid platelet count; however, the response is transitory, and the patients presented a spontaneous regression of thrombocytopenia after 1 to 6 months.16
In consideration of very low platelet count (17 × 109/L), our patient was treated shortly after admission with intravenous γ-globulin of 1 g/kg per day for 2 days. In other case reports, the patients were treated with γ-globulin only for 1 day.13,14 We have preferred 2 administrations in relation to the very low platelet count. The initial response was very good, with a complete regression of purpura on limbs and trunk and an increase in the number of platelets. However, after 1 month of follow-up, we observed a new reduction in the number of platelets. Platelet count returned to normal values after 4 months in relation to undetectable serum CMV DNA and reduction in IgM titer. Also, statistical analysis, in our experience, have demonstrated a significant correlation of platelet count with IgM titer and serum CMV DNA level. It probably demonstrates that intravenous immunoglobulin treatment is not effective in determining a sustained response, but it is useful in cases of purpura or when the platelet level is very low to obtain a rapid, but not sustained, increase of platelet count. Therefore, only the resolution of acute CMV infection determines the complete regression of thrombocytopenia.
In conclusion, this case report shows that, for patients presenting with idiopathic thrombocytopenia, it may be worthwhile to examine for evidence of a recent CMV infection. Intravenous immunoglobulin treatment is effective in determining a rapid increase of platelet count; however, for patients affected with ITP, the response is transitory, and the complete regression of thrombocytopenia is exclusively correlated to the resolution of acute viral infection.
1. Yenicesu I, Yetgin S, Ozyurek E, et al. Virus-associated immune thrombocytopenic purpura in childhood. Pediatr Hematol Oncol. 2002;19(6):433-437.
2. Lazzarotto T, Gabrielli L, Foschini MP, et al. Congenital cytomegalovirus infection in twin pregnancies: viral load in the amniotic fluid and pregnancy outcome. Pediatrics. 2003;112(2):153-157.
3. Liesnard C, Donner C, Brancart F, et al. Prenatal diagnosis of congenital cytomegalovirus infection: prospective study of 237 pregnancies at risk. Obstet Gynecol. 2000;95:881-888.
4. Kylat RI, Kelly EN, Ford-Jones EL. Clinical findings and adverse outcome in neonates with symptomatic congenital cytomegalovirus (SCCMV) infection. Eur J Pediatr. 2006;165(11):773-778.
5. Eisenberg M, Kaplan B. Cytomegalovirus-induced thrombocytopenia in an immunocompetent adult. West J Med. 1993;158:525-526.
6. Wright JG. Severe thrombocytopenia secondary to asymptomatic cytomegalovirus infection in an immunocompetent host. J Clin Pathol. 1992;45:1037-1038.
7. Arruda VR, Rossi CL, Nogueira E, et al. Cytomegalovirus infection as cause of severe thrombocytopenia in a non-immunosuppressed patient. Acta Haematol. 1997;89(4):228-230.
8. Lopez Vivancos J, Falgueras Lopez L, Martin Fuster S, et al. Severe thrombocytopenia secondary to cytomegalovirus infection in an immunocompetent adult. An Med Interna. 1998;15(10):561-562.
9. Noguchi M, Aruga S, Kato A, et al. Cytomegalovirus mononucleosis with severe thrombocytopenia. Rinsho Ketsueki. 2000;41(11):1171-1177.
10. Liberek A, Rytlewska M, Szlagatys-Sidorkiewicz A, et al. Cytomegalovirus disease in neonates and infants-clinical presentation, diagnostic and therapeutic problems-own experience. Med Sci Monit. 2002;8(12):815-820.
11. Bonnet F, Morlat P, Neau D, et al. Hepatologic and immunologic manifestations of primary cytomegalovirus infections in non-immunocompromised hospitalized adults. Rev Med Interne. 2002;21(7):586-594.
12. Flont M. Thrombocytopenia associated with cytomegalovirus infection in an immunocompetent patient. Pol Arch Med Wewn. 2000;104(2):497-499.
13. Nomura K, Matsumoto Y, Kotoura Y, et al. Thrombocytopenia due to cytomegalovirus infection in an immunocompetent adult. Hematology. 2005;10(5):405-406.
14. Alliot C, Barrios M. Cytomegalovirus-induced thrombocytopenia in an immunocompetent adult effectively treated with intravenous immunoglobulin: a case report and review. Hematology. 2005;10(4):277-279.
15. Feng J, Balthasar J. Mechanism of intravenous immunoglobulin action in immune thrombocytopenic purpura. Hum Immunol. 2005;66(4):403-410.
16. Tarantino M. Recent advances in the treatment of childhood immune thrombocytopenic purpura. Semin Hematol. 2006;43:S11-S19.
© 2008 Lippincott Williams & Wilkins, Inc.