The Internet Journal of Internal Medicine™ ISSN: 1528-8382

| Home | Editors | Current Issue | Archives | Instructions for Authors | Disclaimer | Bookmark and Share Share with others |
 

Cytomegalovirus Infection and Thrombosis: Coincidence or Connection?

Lihi Atzmony BSc
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Nao Mashav BSc
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Nili Saar BSc
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Jonathan Canaani MD
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Orit Kliuk-Ben-Bassat MD
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Ora Halutz MD
The Clinical Virology Laboratory, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Yael Paran MD
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Arie Steinvil MD
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Dan Justo MD
Department of Internal Medicine D, Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Israel Physical Address

Citation: L. Atzmony, N. Mashav, N. Saar, J. Canaani, O. Kliuk-Ben-Bassat, O. Halutz, Y. Paran, A. Steinvil & D. Justo : Cytomegalovirus Infection and Thrombosis: Coincidence or Connection?. The Internet Journal of Internal Medicine. 2009 Volume 8 Number 1

Keywords: Cytomegalovirus | Ganciclovir | Thrombosis

 

Abstract

Introduction:Sporadic cases of concomitant acute cytomegalovirus (CMV) infection and thrombosis have been reported several times in the past. However,the incidence of thrombosis among patients with acute CMV infection has never been studied. Moreover, since acute CMV infection is common, it might be diagnosed in patients suffering from thrombosis without actually triggering it.Aim: To study the incidence of thrombosis among patients with acute CMV infection hospitalized in a tertiary medical center. Among these patients, we compare the incidence of CMV-induced thrombosis with that of CMV-accompanied thrombosis.Methods: Medical charts and imaging study reports of all patients diagnosed with acute CMV infection during the years 2005-2006 were reviewed for the presence of thrombosis and for its acquired as well as inherited predisposition.Results: Overall 160 patients were diagnosed with acute CMV infection during the years 2005-2006. Six (3.8%) of these patients were diagnosed concomitantly with thrombosis. Of these patients, four were immunocompromised. Either the acute CMV infection or the thrombosis was diagnosed incidentally in all of these patients. Acquired predispositions for thrombosis, other than acute CMV infection, were present in four of them. Hence, the incidence of CMV-induced thrombosis and CMV-accompanied thrombosis was 1.3% and 2.5%, respectively. Conclusions: Though thrombosis among patients with acute CMV infection is not rare, it is unclear whether CMV actually triggered thrombosis in most of these cases. A case controlled study is needed to confirm the association between acute CMV infection and thrombosis.


Introduction

Concomitant acute cytomegalovirus (CMV) infection and thrombosis has been reported several times in medical literature. Though mainly seen in immunocompromised patients, it has been reported in immunocompetent patients as well [1]. The incidence of thrombosis among immunocompromised patients and immunocompetent patients with acute CMV infection has never been studied, to the best of our knowledge.

In most cases of concomitant acute CMV infection and thrombosis, CMV was presumed to trigger thrombosis, having been diagnosed earlier [1]. Yet, it is doubtful whether acute CMV infection indeed triggered thrombosis in cases where CMV infection was diagnosed following the diagnosis of thrombosis [2]; when triggers for thrombosis, other than acute CMV infection, could be identified [1]; and when the diagnosis of thrombosis was made incidentally [3, 4]. Moreover, since acute CMV infection is quite common [5] and since small thromboses can be diagnosed incidentally, by advanced high-resolution imaging performed for other indications, acute CMV infection might be diagnosed along with thrombosis without actually triggering it.

Aim

We studied the incidence of thrombosis among immunocompromised and immunocompetent patients with acute CMV infection hospitalized in a tertiary medical center. Among these patients, we compared the incidence of CMV-induced thrombosis with that of CMV-accompanied thrombosis.

Methods

Inclusion criteria

The study design was that of an observational retrospective approach. We reviewed records of the clinical virology laboratory at the Tel-Aviv Sourasky Medical Center (TASMC) from the years 2005-2006 in search of all patients with acute CMV infection. Following that, all medical charts and imaging study reports of these patients were retrieved from the Sourasky Medical Center archive and reviewed for any thrombosis findings, along with clinical aspects, and acquired as well as inherited predisposition information. The study was approved by the local ethics committee.

Definition of acute CMV infection

Patients with acute CMV infection were defined as those whose serum was found positive for CMV-specific IgM antibodies (VIDAS enzyme-linked fluorescent assay, Bio Mérieux, Marcy-l’Etoile, France) with or without evidence for serum conversion, i.e., previous CMV-specific IgG antibodies in serum, and/or those positive for CMV pp65 antigens on peripheral blood leukocytes (Chemicon International Inc., Temecula, Calif.). Clinical aspects of the CMV infection were also documented; including signs of infectious mononucleosis, i.e., lymphocytosis, monocytosis, elevated liver enzymes, pneumonitis, retinitis, colitis, and any other organ CMV infection might involve [5].

Definition of thrombosis

Since CMV-specific IgM antibodies in serum might not be detected for up to four weeks following CMV infection, and since titers might have remained high for months following infection [5], we looked for thrombosis one month previous to the diagnosis of CMV infection and up to three months following it. Included were patients with venous thrombosis as well as patients with arterial thromboembolism, provided they did not suffer from atrial fibrillation or infective endocarditis. Also included, were patients presenting indirect evidence for thrombosis, such as a splenic infract. Patients with acute coronary syndromes and patients with cerebrovascular events were excluded from this study.

Thrombosis predisposition

All medical charts of patients with acute CMV infection and thrombosis were reviewed for acquired as well as inherited predisposition for thrombosis, including prolonged air-travel, prolonged immobility, obesity, use of oral contraceptives, pregnancy, postmenopausal hormone replacement therapy, recent surgery, trauma, indwelling intravenous lines, medical conditions such as antiphospholipid antibody syndrome, history of previous thrombosis, family history of thrombosis and genetic thrombophilias, such as factor V Leiden, protein S deficiency, and protein C deficiency [6].

CMV-induced thrombosis versus CMV-accompanied thrombosis

Patients with CMV-induced thrombosis were those with no predispositions for thrombosis, other than acute CMV infection. Patients with CMV-accompanied thrombosis were those with predispositions for thrombosis, other than acute CMV infection.

Results

A total of 160 patients were diagnosed with acute CMV infection during the years 2005-2006 at the TASMC. Six (3.8%) of these patients, 2 male and 4 female (mean age 45.8±12.5 years, median age 47 years), experienced thrombosis during acute CMV infection (table 1). Four of these were immunocompromised (i.e. transplant recipients, and patients taking systemic steroids). Four had lab results consistent with acute CMV infection (lymphocytosis, monocytosis, elevated liver enzymes). Acute CMV infection and/or thrombosis were diagnosed incidentally in all of these patients without actually suspecting either of them. Acute CMV infection was diagnosed in all transplant recipients through CMV tests performed routinely in case of fever in these patients. Four patients had some form of predisposition for thrombosis, i.e. recent operation and use of oral contraceptives. Only one patient had gone through genetic hypercoagulability investigation, which was negative. The incidence of CMV-induced thrombosis (n=2) and CMV-accompanied thrombosis (n=4) was 1.3% and 2.5%, respectively.

Case histories

Patient 1

A 54-year-old male with a six week history of fever was hospitalized in September 2006. His medical history was consistent with bronchial asthma. Physical examination revealed oral temperature of 38.4˚C. His complete blood count was consistent with lymphocytosis and monocytosis (table). In the course of investigating fever of unknown origin, spleen infarcts were demonstrated by abdominal computed tomography. No vegetations or mural thrombus were found on transesophageal echo. Blood cultures were sterile. No EBV-IgM antibodies were found in the serum, while serological testing for CMV-IgM antibodies was positive. The patient was discharged with no treatment.

Patient 2

A 58-year-old female was hospitalized in October 2006 with a six day history of right arm swelling around a hemodialysis arterial-venous fistula. She had undergone kidney transplantation six years previously, and had been taking immunosuppressive agents. She also had a history of atrial fibrillation but had not been treated with anticoagulation. Physical examination revealed a painful, hyperemic right arm. Her complete blood count and liver enzymes were within normal range (table). A brachial DVT was demonstrated by Doppler ultrasonography. Serological testing for CMV-IgM antibodies, performed as part of the routine investigation of fever in a transplant recipient, was positive. The patient was treated with Valciclovir.

Patient 3

A 29-year-old female with chest pain, dyspnea and low-grade fever was hospitalized in September 2005. She used oral contraceptives. Physical examination revealed normal breath sounds. Chest roentgenogram was also normal. Bilateral pulmonary emboli were diagnosed by chest computed tomography angiography. No DVT was demonstrated by Doppler ultrasonography of the lower extremities. The patients fever continued over the following days. Physical examination was unremarkable. Complete blood count showed mild lymphocytosis (table). Blood cultures were sterile. Serological testing for CMV-IgM antibodies was positive twice. Hypercoagulability workup was unremarkable. The patient was discharged with no treatment other than anticoagulant treatment.

Patient 4

A 36-year-old female was hospitalized in February 2006 due to fever few hours following dialysis. Her medical history consisted of systemic lupus erythematotus, end-stage renal disease, and severe peripheral vascular disease. She was treated with systemic steroids due to systemic lupus erythematotus. The patient was discharged from hospitalization 5 days prior to current admission, following a 4 months hospitalization period, during which she underwent colectomy due to ischemic colitis, following an aorto-femoral bypass surgery. About a month prior to her current admission, during this previous episode, a wedge-shaped spleen infarct was demonstrated on an abdominal CT. Physical examination during current hospitalization revealed drowsiness, an oral temperature of 380C, and very low blood pressures. Liver enzymes were normal (table). As a state of sepsis was suspected, a wide range antibiotic treatment was initiated. A few days later, a CMV PP65 antigenemia assay showed acute CMV infection. Abdominal CT demonstrated new bilateral hypodensic areas in the patient's kidneys, which could correlate to kidney infarcts. No vegetations or mural thrombus were found on transesophageal echo. Blood cultures were sterile. Upon initiating treatment with Ganciclovir the patient's fever and CMV antigenemia levels decreased.

Patient 5

A 58-year-old male was hospitalized in November 2005 with fever. His medical history consisted of liver transplantation 3 months prior to hospitalization due to hepatocellular carcinoma, ischemic heart disease, diabetes mellitus, chronic renal failure and hypertension. Physical examination revealed lower extremities edema. His complete blood count was consistent with thrombocytopenia, and liver enzymes were elevated (table). Due to a positive CMV PP65 antigenemia assay, performed as part of the routine investigation of fever in a transplant recipient, IV Ganciclovir treatment was initiated through a central catheter in the patient's right arm, and fever resolved few days later. An additional CMV PP65 antigenemia assay, performed following 3 weeks of treatment with IV Ganciclovir, was negative, and the catheter was removed. The following day, the patient started complaining of pain and swelling in his right arm. Physical examination revealed a painful hyperemic right arm. A right brachial DVT was demonstrated by Doppler ultrasonography, and the patient was started on anticoagulant treatment.

Patient 6

A 40-year-old female was hospitalized in August 2006 due to low grade fever and diarrhea lasting 3 days. Her medical history consisted of liver transplantation two months earlier, due to alcoholic cirrhosis, and current immunosuppressive treatment. Physical examination revealed an oral temperature of 38.2˚C, and hepato-splenomegaly. Lab results showed thrombocytopenia (table). Abdominal CT showed evidence of portal vein thrombosis, and a CMV antigenemia assay, performed few days later as part of the routine investigation of fever in a transplant recipient, was positive. IV Ganciclovir as well as anticoagulant treatment was started.

Thumbnail: Table 1: Laboratory findings of patients with CMV infection and thrombosis
Table 1: Laboratory findings of patients with CMV infection and thrombosis

PT = Prothrombin Time; PTT = Partial Prothrombin Time; WBC = White Blood Count; PLT = Platelets; ALT = Alanine Transferase; AST = Aspartate Transferase; ALP = Alkaline Phosphatase; GGT = Gamma Glutamyl Transferase; LDH = Lactate Dehydrogenase

Discussion

Acute CMV infection is quite common, and positive serology for CMV ranges from 40% to 100% worldwide [5]. Concomitant acute CMV infection and thrombosis has been reported several times in medical literature, mainly in immunocompromised patients, such as HIV-infected patients [1]. Thrombosis has also been documented after solid-organ transplantations, and a causal relationship has been shown between acute CMV infection and the rejection of renal and liver allografts, suspected to be mediated by thrombosis [7, 8]. Reports on concomitant acute CMV infection and thrombosis in immunocompetent patients are mostly sporadic [1], though Fridlenger et al. have recently described a series of such patients [9]. The incidence of thrombosis among immunocompromised and immunocompetent patients with acute CMV infection has never been studied. In our study, we presented six patients with concomitant acute CMV infection and a thrombotic event, in a two year retrospective cohort of 160 patients admitted to a tertiary hospital and diagnosed with CMV. Of these six patients, two were immunocompetent, and the remaining four, immunocompromised. In all of these patients, acute CMV infection and/or thrombosis was diagnosed incidentally in the process of investigating fever of unknown origin, or during routine testing for CMV as part of fever work-up in transplanted patients. Most patients had acquired predispositions for thrombosis other than acute CMV infection. Due to accumulating evidence that a relationship between acute CMV infection and thrombosis does indeed exist, we think our data should encourage further research on the co-incidence of CMV and thrombosis, a topic that undoubtfully deserves further investigation, in order to raise awareness to this unfamiliar CMV infection outcome, which could affect current anticoagulation treatment guidelines.

The exact pathological mechanism underlying the relationship between acute CMV infection and thrombosis is currently unclear. Current theories suggest that CMV infection could trigger thrombosis by enhancing platelet and leukocyte adhesion to infected endothelial cells, or, alternatively, by increasing the circulatory levels of factor VIII. Other theories pose that CMV induces transient antiphospholipid antibody production, or that it enhances vascular smooth muscle proliferation [10-14].

We acknowledge the major limitation of our current study being its retrospective observational design, lacking a control group of patients with no CMV infection, utilizing data from a single hospital, perhaps misrepresenting true incidence of thrombotic events in acute CMV infection. However, with no prior similar studies available, we feel our findings could augment future comparative studies. We thus conclude that a prospective case-controlled study on this significant complication of acute CMV infection is highly warranted.

Correspondence to

Dr. Dan Justo, Department of Internal Medicine D, Sourasky Medical Center, 6 Weitzman Street, Tel-Aviv 64239, Israel, Tel: 972-52-4266739, Fax: 972-9-7408575, e-mail: justo1@bezeqint.net

References

1. Squizzato A, Gerdes VE, Buller HR. Effects of human cytomegalovirus infection on the coagulation system. Thromb Haemost 2005;93:403-10. (s)

2. Paran Y, Halutz O, Swartzon M, Schein Y, Yushurun D, Justo D. Venous thromboembolism and cytomegalovirus infection in immunocompetent adults. Isr Med Assoc J 2007;9:757-758. (s)

3. Ailani RK, Simms R, Caracioni AA, West BC. Extensive mesenteric inflammatory veno-occlusive disease of unknown etiology after primary cytomegalovirus infection: first case. Am J Gastroenterol 1997;92:1216-8. (s)

4. Garcia I, Fainstein V, Rios A, Luna M, Mansell P, Reuben J, Hersh E. Nonbacterial thrombotic endocarditis in a male homosexual with Kaposi's sarcoma. Arch Intern Med 1983;143:1243-4. (s)

5. Taylor GH. Cytomegalovirus. Am Fam Physician 2003;67:519-24. (s)

6. Heit JA, O’Fallon WM, Petterson TM. Relative impact of risk factors for deep vein thrombosis and pulmonary embolism: a population-based study. Arch Intern Med 2002;162:1245–48. (s)

7. Madalosso C, de Souza NF Jr, Ilstrup DM, Wiesner RH, Krom RA. Cytomegalovirus and its association with hepatic artery thrombosis after liver transplantation. Transplantation 1998;66:294-7. (s)

8. Kazory A, Ducloux D, Coaquette A, Manzoni P, Chalopin JM. Cytomegalovirus-associated venous thromboembolism in renal transplant recipients: a report of 7 cases. Transplantation 2004;77:597-9. (s)

9. Fridlender ZG, Khamaisi M, Leitersdorf E. Association between cytomegalovirus infection and venous thromboembolism. Am J Med Sci 2007;334:111-4. (s)

10. Span AH, van Dam-Mieras MC, Mullers W, Endert J, Muller AD, Bruggeman CA. The effect of virus infection on the adherence of leukocytes or platelets to endothelial cells. Eur J Clin Invest 1991;21:331-8. (s)

11. Uthman IW, Gharavi AE. Viral infections and antiphospholipid antibodies. Semin Arthritis Rheum 2002;31:256-63. (s)

12. Schambeck CM, Hinney K, Gleixner J, Keller F. Venous thromboembolism and associated high plasma factor VIII levels: linked to cytomegalovirus infection? Thromb Haemost 2000;83:510-1. (s)

13. Delbos V, Abgueguen P, Chenneebalut JM. Acute Cytomegalovirus infection and venous thrombosis: role of antiphospholipid andibodies. J infect 2007;54:e47-50. (s)

14. Zhou YF, Yu ZX, Wanishsawad C. The immediate early genome products of of human cytomegalovirus increase vascular smooth muscle cell migration, proliferation and expression of PDGF beta-receptor. Biochem Biophys Res Commun 1999;256:608-13. (s)

This article was last modified on Sat, 07 Nov 09 19:36:15 -0600

This page was generated on Wed, 30 Dec 09 12:31:04 -0600, and may be cached.
Advertisement