The antiphospholipid antibody positivity (aPL) and antiphospholipid syndrome (APS) can be associated with a transient or permanent infectious insight [1], and CMV infection is a rare association, which has been sporadically reported in the literature [2]. APS diagnosis was established by the combination of at least one clinical criteria and one analytical, according to the Sydney classification criteria [3]. The clinical criteria comprise the presence of arterial thrombosis, venous or small vessel, or the presence of obstetric morbidity. Laboratory criteria include: a) presence of lupus anticoagulant (LA); b) medium or high titter (or>99% percentile) IgG/IgM anticardiolipin antibodies (aCL); or c) IgG/IgM antibeta2glycoprotein I antibodies (anti-β2GPI) (or>99 % percentile). These autoantibodies must remain positive at least 12 weeks [3]. The pro-thrombotic risk due to positive aPL-related infection is a matter of discussion, since it seems these aPL are functionally inefficient. The JAK2 V617F mutation is an acquired disorder related to chronic myeloproliferative disorders (MPD). There is increasing clinical evidence to suggest that the JAK2 mutation may be variably associated with thrombosis [12,13]. The role of screening for the JAK2 V617F mutation in patients presenting with thrombosis without overt MPD is unclear, but it appears justified in cases of idiopathic splanchnic vein thrombosis [4]. Herein, we communicate a challenging case of a previously healthy young man with multiple visceral thrombosis and three different acquired thrombophilic risk factors, aPL laboratory category I included.

A 39 years healthy man was admitted to the emergency department of our hospital with the diagnosis of fever and splenomegaly. The patient complained of a ten day pain in the right upper quadrant. In the two weeks prior to his admission, he developed flu-like joint pain symptoms, fever, severe fatigue and dry cough.

At the hospital, an abdominal CT scan showed severe splenomegaly, with two triangular peripheral images with filling defects (one in medial region of 20 × 24 × 20 mm and bottom of 8 × 11 × 8 mm) compatible with splenic infarction (Figure 1). No remarkable lymphadenopathy was observed and other structures remained unaltered. Differential diagnosis of haematological disease, congenital or acquired thrombophilia and embolic disease was raised. The study was completed with a transthoracic echocardiography, being completely normal. Chest CT scan showed multiple cavitated and diffuse pulmonary micro nodules, predominantly in the upper lobes related to small peripheral arterial branches (Figure 1). The patient had a blood test without noticeable changes except for a striking alteration in partial activated thromboplastin time (APTT) of 42 seconds (r=1.45) (level: 24-36; r ≤ 1.2). We observed normal liver and kidney function, with minimally elevated LDH 447 U/L (208-378 U/L), and erythrocyte sedimentation rate andrheumatoid factor in the normal range. Tumour markers showed no alterations. PPD and booster were negative. Serology was negative for HIV, HBV, HCV, EBV, B19 Parvovirus, Mycoplasma, and Chlamydia pneumoniae, Legionella, Coxiella, Toxoplasma, Brucella and Leishmania.

Figure 1: Abdominal CT scan showed severe splenomegaly, with two triangular peripheral images with filling defects. Chest CT scan showed multiple cavitated and diffuse pulmonary micro nodules.

A positive value for anti-CMV IgM was obtained and persisted positive two weeks later. Finally, after 16 weeks the values became negative, and anti-CMV IgG turned positive (Figure 2). In addition to this, an autoimmune study was performed, and ANAs and ANCA were determined being both negative. Due to the enlargement of APTT, a study to acquired thrombophilia was performed, obtaining a positive LA, negative IgM and IgG anti-beta2GPI, and low positivity titters of IgM aCL (30 MPL/ml with normal values <20 MPL/mL), together with negative aCL-IgG. Sixteen weeks later, IgM-aCL turned negative (Figure 2). LA positivity persisted at first time and later became negative. Inherited thrombophilia was checked at the acute phase with both a genetic test for paroxysmal nocturnal haemoglobinuria (PNH) and JAK2, since other tests have little value at this point. The study was negative for PNH, but positive for JAK2, twice confirmed. An aspirate and bone marrow biopsy (BMB) were done, with no cytogenetic abnormalities to show. In a second phase, antithrombin, thrombin time, free protein S, protein C, total fibrinogen, resistance to activated protein C, FII – 20210A – and factor V Leiden mutations, factor VIII and homocysteine were tested, being all normal.

Figure 2: A positive value for anti-CMV IgM was obtained and persisted positive two weeks later. Finally, at 16 weeks the values became negative, and anti-CMV IgG turned positive. A positive LA, low positivity titters of IgM-aCL, together with negative aCL-IgG was obtained on admission. Sixteen weeks later, IgM-aCL turned negative. LA positivity persisted at first time and later became negative.

The patient was initially treated with therapeutic low molecular weight heparin (LMWH) dose, shifting to acenocumarol once finished all invasive diagnostic tests. He was not treated by specific therapy for CMV infection due to the absence of demonstrated immunosuppression. A control thoracic-abdominal CT scan showed a decrease in the number of the pulmonary images with persistent splenomegaly and splenic infarction. We consider that the patient should be anticoagulated for long time being monitored in outpatient routine.

It has been reported that acute infections can become a risk factor for the development of thrombotic events, many of them being favoured by pre-existing silent thrombophilia [5], as happened in our case. Thus, an incidence of 6.4% of thrombosis in patients with acute CMV infection has been reported. This fact seems in contradiction with the belief that a patient will have more thrombotic complications if there is a certain degree of immunosuppression. A recent metaanalysis [6] proved that 1/3 of patients with thrombotic events and acute CMV infection were previously immunosuppressed while the remaining 2/3 were not immunosuppressed. There are several mechanisms caused by CMV infection that could lead to a prothrombotic state, but we wish to emphasize the ones related to the APS [7]. It has been hypothesized that molecular mimicry between viral particle and endothelium generates the mentioned antibodies. A study demonstrated that aPL induced in mice by immunization with a cytomegalovirus derived peptide is pathogenic in vivo, inducing proinflammatory and pro-coagulant state [8].

Although aCL-ELISA is not aimed at detecting β2-GPI, sometimes those can be detected in the presence of concomitant infections. On the other hand, anti-β2GPI antibodies can be present in cases of leishmaniosis or leprosy [9,10]. A key differential feature between anti- β2GPI antibodies occurring in the context of the former infectious diseases and those that strongly associate with thrombosis refers to the different domain triggered in the β2-GPI molecule (domain I instead of V) [9]. In addition, anti-β2GPI antibodies associated with APS are predominantly the IgG isotype, particularly IgG2 subclass, whereas those occurring in association with infections are of the IgM isotype. Interestingly, the LA assay detecting either anti-β2GPI or antiprothrombin antibodies can also be positive also with concurrent infections.

The “two hit” theory of thrombosis associated with the APS stated that an initiating “first hit” injury disrupts the endothelium, and a “second hit” activates the clot formation [11]. Thus, we could hypothesize that CMV provoked an endothelial injury that temporally facilitated the apparition of aPL, LA and aCL-IgM in our case, in a similar form to what occurs in catastrophic APS. The concurrence of aCL-IgM isotype with LA positivity due to anti-β2GPI or maybe to antiprothrombin antibodies reinforces our hypothesis that these aPL, although transient, could be able to act as a trigger of the clot activation pathway, particularly in a carrier of JAK2 mutation.

JAK2 mutation has been described in some myeloproliferative neoplasms that are frequently associated with arterial thrombosis, venous thrombosis andthromboembolism [12]. JAK2 V617F induces biological changes to megakaryocytes and platelets leading to increase thrombotic events. Specifically it amplifies the signal through the thrombopoietin (TPO)/MPL pathway in megakaryocytes, increasing formation of proplatelets; and also increasing aggregation, spreading, and thrombus formation of platelets [13]. We want to clarify that idiopathic visceral thrombosis patients should be screened from thrombophilic risk factors, including infections and the JAK2 V617F mutation. At least, we should notice that patients with JAK2 mutation are more likely to develop a myeloproliferative neoplasms that patients with another concomitant mutation. We think that it could be related with the demonstrate mutation order in these disorders. Patients who acquire a JAK2 mutation first produce excess differentiated cells in the absence of a distinct self-renewal advantage in the hematopoietic stem cells. Because of that monitoring these patients should be warranted [14].

Authors are grateful to Josep Gonzalez and Elena Cabezudo. Thanks to Mercè Bacardit for the English correction.